Igor_Grebeniuk_final.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b3256aca-1544-492a-bb5a-58ffb4355592",
   "metadata": {},
   "source": [
    "# **Instructions:**\n",
    "There are no correct or wrong answers to this assignment.\n",
    "Please immerse yourself into the role of an Anti-Fraud Analyst (whatever it means for you), and demonstrate via the use of code and comments\n",
    "your abilities and approaches to data exploration, modeling, and investigation. You may write a summary of your findings at the end of your\n",
    "file submission, as well as proposals to any (if there are any) fraudulent patterns you are able to detect.\n",
    "In the event that you have any follow up questions or need for clarifications, feel free to contact the email address from which you have \n",
    "received this assignment."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a1da8968-691d-47a0-9bdc-8e0e11e16d22",
   "metadata": {},
   "source": [
    "Please find attached 2 csv files:\n",
    "AntiFraudAnalystTradeData.csv - contains all trades of a specific time period, from the point of the view of the user (i.e. positive pnl = user profit)\n",
    "AntiFraudAnalystUserDevices.csv - contains all available device ids of users who traded in this time period"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "189b20e5-a900-4460-b236-464a28463bb8",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4a06ac33-82bc-4b56-9e6a-127bb6742178",
   "metadata": {},
   "outputs": [],
   "source": [
    "import datetime\n",
    "from datetime import datetime"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "96ade4dc-1c01-4eda-98cb-8731ffa0ef96",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "plt.rcParams['figure.figsize'] = (5, 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "cf04a13a-1ec2-452e-a913-0a5081584bf3",
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "029887f0-4e04-4fcf-88ba-d7de3ff2f7cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_trade = pd.read_csv('AntiFraudAnalystTradeData.csv')\n",
    "df_ids = pd.read_csv('AntiFraudAnalystUserDevices.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf376b46-14e0-4be6-bff5-a787211173a9",
   "metadata": {},
   "source": [
    "#### **Columns:**  \n",
    "id - unique id of the trade  \n",
    "user_id - unique id of the user  \n",
    "registration_date - registration date of the user  \n",
    "fin_instrument - financial asset name of the underlying  \n",
    "instrument_type - type of option instrument  \n",
    "instrument_direction - put/call direction of the trade  \n",
    "create_at - time of trade open  \n",
    "close_at - time of trade close  \n",
    "instrument_expiration - maturity time of option product  \n",
    "close_reason - reason for closing the position  \n",
    "platform - type of device platform trade was executed on  \n",
    "open_quote - market quote at time of open  \n",
    "close_quote - market quote at time of close  \n",
    "instrument_strike - strike of the option instrument  \n",
    "initial_investment_enrolled - dollar amount of investment  \n",
    "pnl_total_enrolled - dollar amount of profit/loss  \n",
    "#------------------------------------------------------------  \n",
    "user_id - unique id of the user  \n",
    "device_id - unique id of the user's device"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "05f11c68-f891-4e4d-a246-ce09a6d6296f",
   "metadata": {},
   "source": [
    "## **Plan to resolve the task:**  "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb76b933-54da-4671-80f7-3107a5061d35",
   "metadata": {},
   "source": [
    "1. Run some analysis (EDA) for data.\n",
    "2. ML modeling: Re-format and fill in with new features the data with trades with the focus on following features which may be crucial to determine fraud:  \n",
    "   a) number of devices per user,  \n",
    "   b) accounts using same devices,  \n",
    "   c) number of trades by user,  \n",
    "   d) total pnl by user,  \n",
    "   e) count of calls and puts by user,  \n",
    "   f) count of option types (turbo, digital) by user,  \n",
    "   g) count of platforms used by a user,  \n",
    "   h) count of wins vs. losses by a user,  \n",
    "   i) total investment by a user,  \n",
    "   j) estimation of activity level as per trades per time period,  \n",
    "   k) registration year,\n",
    "   l) number of questionable trades with pnl above or below some criteria,  \n",
    "   m) number of questionable trades with pnl above or below some criteria executed at \"silence\" period when trading level is low.  \n",
    "3. Run Cluster analysis and make conclusions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5627bcb3-e449-411b-948b-5b6ccb06ee0e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>user_id</th>\n",
       "      <th>registration_date</th>\n",
       "      <th>fin_instrument</th>\n",
       "      <th>instrument_type</th>\n",
       "      <th>instrument_direction</th>\n",
       "      <th>create_at</th>\n",
       "      <th>close_at</th>\n",
       "      <th>instrument_expiration</th>\n",
       "      <th>close_reason</th>\n",
       "      <th>platform</th>\n",
       "      <th>open_quote</th>\n",
       "      <th>close_quote</th>\n",
       "      <th>instrument_strike</th>\n",
       "      <th>initial_investment_enrolled</th>\n",
       "      <th>pnl_total_enrolled</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>9653594472</td>\n",
       "      <td>585118157</td>\n",
       "      <td>2019-04-05 17:12:46.000000</td>\n",
       "      <td>XRPUSD</td>\n",
       "      <td>turbo-option</td>\n",
       "      <td>call</td>\n",
       "      <td>2022-06-17 19:41:03.216000+00:00</td>\n",
       "      <td>2022-06-17 19:42:00+00:00</td>\n",
       "      <td>2022-06-17 19:42:00+00:00</td>\n",
       "      <td>expired</td>\n",
       "      <td>Unknown platform</td>\n",
       "      <td>0.314705</td>\n",
       "      <td>0.314955</td>\n",
       "      <td>0.314705</td>\n",
       "      <td>22.61</td>\n",
       "      <td>19.21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>9641975390</td>\n",
       "      <td>1456695825</td>\n",
       "      <td>2021-08-12 02:28:57.928101</td>\n",
       "      <td>XRPUSD</td>\n",
       "      <td>turbo-option</td>\n",
       "      <td>call</td>\n",
       "      <td>2022-06-14 00:23:15.137000+00:00</td>\n",
       "      <td>2022-06-14 00:24:00+00:00</td>\n",
       "      <td>2022-06-14 00:24:00+00:00</td>\n",
       "      <td>expired</td>\n",
       "      <td>WebGL</td>\n",
       "      <td>0.308125</td>\n",
       "      <td>0.307895</td>\n",
       "      <td>0.308125</td>\n",
       "      <td>0.39</td>\n",
       "      <td>-0.39</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0          id     user_id           registration_date  \\\n",
       "0           0  9653594472   585118157  2019-04-05 17:12:46.000000   \n",
       "1           1  9641975390  1456695825  2021-08-12 02:28:57.928101   \n",
       "\n",
       "  fin_instrument instrument_type instrument_direction  \\\n",
       "0         XRPUSD    turbo-option                 call   \n",
       "1         XRPUSD    turbo-option                 call   \n",
       "\n",
       "                          create_at                   close_at  \\\n",
       "0  2022-06-17 19:41:03.216000+00:00  2022-06-17 19:42:00+00:00   \n",
       "1  2022-06-14 00:23:15.137000+00:00  2022-06-14 00:24:00+00:00   \n",
       "\n",
       "       instrument_expiration close_reason          platform  open_quote  \\\n",
       "0  2022-06-17 19:42:00+00:00      expired  Unknown platform    0.314705   \n",
       "1  2022-06-14 00:24:00+00:00      expired             WebGL    0.308125   \n",
       "\n",
       "   close_quote  instrument_strike  initial_investment_enrolled  \\\n",
       "0     0.314955           0.314705                        22.61   \n",
       "1     0.307895           0.308125                         0.39   \n",
       "\n",
       "   pnl_total_enrolled  \n",
       "0               19.21  \n",
       "1               -0.39  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_trade.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c857917b-3604-42df-9a54-6f9746795703",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>user_id</th>\n",
       "      <th>device_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1185175017</td>\n",
       "      <td>a3037561a43a726fdbd4168b4e2e7df7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1245468614</td>\n",
       "      <td>92e5e634-9d56-4aa9-9dd3-9cca03da117e</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0     user_id                             device_id\n",
       "0           0  1185175017      a3037561a43a726fdbd4168b4e2e7df7\n",
       "1           1  1245468614  92e5e634-9d56-4aa9-9dd3-9cca03da117e"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_ids.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "0471a424-3c9e-420c-a28c-810ba93244a7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1764494 entries, 0 to 1764493\n",
      "Data columns (total 17 columns):\n",
      " #   Column                       Dtype  \n",
      "---  ------                       -----  \n",
      " 0   Unnamed: 0                   int64  \n",
      " 1   id                           int64  \n",
      " 2   user_id                      int64  \n",
      " 3   registration_date            object \n",
      " 4   fin_instrument               object \n",
      " 5   instrument_type              object \n",
      " 6   instrument_direction         object \n",
      " 7   create_at                    object \n",
      " 8   close_at                     object \n",
      " 9   instrument_expiration        object \n",
      " 10  close_reason                 object \n",
      " 11  platform                     object \n",
      " 12  open_quote                   float64\n",
      " 13  close_quote                  float64\n",
      " 14  instrument_strike            float64\n",
      " 15  initial_investment_enrolled  float64\n",
      " 16  pnl_total_enrolled           float64\n",
      "dtypes: float64(5), int64(3), object(9)\n",
      "memory usage: 228.9+ MB\n"
     ]
    }
   ],
   "source": [
    "df_trade.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "1b42aab5-04b0-406d-9505-25de22d8f1be",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 6245842 entries, 0 to 6245841\n",
      "Data columns (total 3 columns):\n",
      " #   Column      Dtype \n",
      "---  ------      ----- \n",
      " 0   Unnamed: 0  int64 \n",
      " 1   user_id     int64 \n",
      " 2   device_id   object\n",
      "dtypes: int64(2), object(1)\n",
      "memory usage: 143.0+ MB\n"
     ]
    }
   ],
   "source": [
    "df_ids.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "09d01e94-d030-490a-9fe1-b41d756566e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# as 'Unnamed: 0' columns just repeat indexes let's remove them \n",
    "df_trade.drop('Unnamed: 0', axis = 1, inplace = True)\n",
    "df_ids.drop('Unnamed: 0', axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "0f95135a-d6b8-453d-805c-dad5bf420eb5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id                             0\n",
       "user_id                        0\n",
       "registration_date              0\n",
       "fin_instrument                 0\n",
       "instrument_type                0\n",
       "instrument_direction           0\n",
       "create_at                      0\n",
       "close_at                       0\n",
       "instrument_expiration          0\n",
       "close_reason                   0\n",
       "platform                       0\n",
       "open_quote                     0\n",
       "close_quote                    0\n",
       "instrument_strike              0\n",
       "initial_investment_enrolled    0\n",
       "pnl_total_enrolled             0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_trade.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "574f667c-0953-40e1-9c74-e21c60067551",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "user_id      0\n",
       "device_id    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_ids.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "afc7d429-e2ca-43d9-8b54-ef1385172931",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_trade['registration_date'] = pd.to_datetime(df_trade['registration_date'])\n",
    "df_trade['close_at'] = pd.to_datetime(df_trade['close_at'], format='mixed')\n",
    "df_trade['create_at'] = pd.to_datetime(df_trade['create_at'], format='mixed')\n",
    "df_trade['instrument_expiration'] = pd.to_datetime(df_trade['instrument_expiration'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "b8307547-27bd-4efa-9eb0-c46fa70f36db",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1764494 entries, 0 to 1764493\n",
      "Data columns (total 16 columns):\n",
      " #   Column                       Dtype              \n",
      "---  ------                       -----              \n",
      " 0   id                           int64              \n",
      " 1   user_id                      int64              \n",
      " 2   registration_date            datetime64[ns]     \n",
      " 3   fin_instrument               object             \n",
      " 4   instrument_type              object             \n",
      " 5   instrument_direction         object             \n",
      " 6   create_at                    datetime64[ns, UTC]\n",
      " 7   close_at                     datetime64[ns, UTC]\n",
      " 8   instrument_expiration        datetime64[ns, UTC]\n",
      " 9   close_reason                 object             \n",
      " 10  platform                     object             \n",
      " 11  open_quote                   float64            \n",
      " 12  close_quote                  float64            \n",
      " 13  instrument_strike            float64            \n",
      " 14  initial_investment_enrolled  float64            \n",
      " 15  pnl_total_enrolled           float64            \n",
      "dtypes: datetime64[ns, UTC](3), datetime64[ns](1), float64(5), int64(2), object(5)\n",
      "memory usage: 215.4+ MB\n"
     ]
    }
   ],
   "source": [
    "df_trade.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7540fb0f-d2e6-4d65-a79d-71dce8900e55",
   "metadata": {},
   "source": [
    "Let's investigate the number of unique user_ids in both sets ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "fc96fac6-7ba5-4bdb-af9e-41284c13c632",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "81726"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_trade['user_id'].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "f7330f50-fb8b-4708-b68a-6f5fa5a1eeab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "81006"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_ids['user_id'].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f59a630f-6fc2-4044-8fb1-3498732aa886",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "720"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(list(set(list(df_trade['user_id'].unique()))-set(list(df_ids['user_id'].unique()))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "8e911160-2d29-4c51-94d4-cf3b5f66f0d3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(list(set(list(df_ids['user_id'].unique()))-set(list(df_trade['user_id'].unique()))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "bb4146fb-cd23-4dd7-abcb-0d62c52c00a0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.88"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(720/81726*100,2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "282e4d67-331e-48c9-a1ac-fc29f1c6dba4",
   "metadata": {},
   "source": [
    "##### **Finding:**  \n",
    "in df_trade there are more unique user_ids and 720 users in df_trade will be without the total number of used devices as I plan to join two datasets. From another side, 720 is less than 1% of total number of users, so the records for the users may be deleted if no significant impact on overall results will happen. To stay with those users I will equalize number of devices to 0s for those 720 users id. Then, more investigation will be executed."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea0413da-bf1a-4d86-b364-582dcf3a533d",
   "metadata": {},
   "source": [
    "### Let's investigate df_ids dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "890dede0-e832-44bf-810a-1de24c73ed78",
   "metadata": {},
   "outputs": [],
   "source": [
    "# number of devices per a user\n",
    "n_devices_per_user = df_ids.groupby('user_id', as_index=False).count().sort_values(\n",
    "    'device_id', ascending=False).rename(columns = {'device_id':'n_devices'})\n",
    "n_devices_per_user.reset_index(drop = True, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "6c8de29f-ed0a-47c9-9156-a485a072ff7e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>n_devices</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>739862968</td>\n",
       "      <td>17359</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1263693067</td>\n",
       "      <td>15279</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>583977446</td>\n",
       "      <td>14004</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      user_id  n_devices\n",
       "0   739862968      17359\n",
       "1  1263693067      15279\n",
       "2   583977446      14004"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n_devices_per_user.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "44af9084-ff8c-4d47-8607-d4b6e7485619",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# there are some outliers above 2500 n_devices  /user\n",
    "sns.lmplot(data = n_devices_per_user, x = 'user_id', y = 'n_devices', height = 3, aspect =2)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "da31a708-61de-4601-9000-becd7922c608",
   "metadata": {},
   "source": [
    "#### Visually, it looks 2500 number divides user_id outliers from the most of user_ids. let's calculate the share of devices at outliers vs. the total number of devices."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "9a3e6d83-4640-41ed-b214-86f70788aded",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9.9"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(100*n_devices_per_user.query('n_devices > 2500')['n_devices'].sum()/n_devices_per_user['n_devices'].sum(),1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "c360a4ef-5b47-4cf9-a9ae-7dbb8cd3c2e9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "146"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n_devices_per_user.query('n_devices > 2500')['n_devices'].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "91d88218-2de0-4f81-b147-fd757c9583ba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(100*n_devices_per_user.query('n_devices > 2500')['n_devices'].count()/n_devices_per_user['n_devices'].count(),1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f03c468d-c26f-4549-afc2-0da04b140586",
   "metadata": {},
   "source": [
    "**Findings:**  \n",
    "a) The impact on number of devices from those outliers is 10% vs. the total number  \n",
    "b) But, the impact is provided by just 146 users that is only 0.2% of all users in the dataset.\n",
    "\n",
    "Those users are not physical people but companies providing the service or this is sign of fraud?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "541c2f2a-975a-4587-9f8c-d9ad56f7cb0a",
   "metadata": {},
   "source": [
    "#### Let's analyze how many users use 1, 2, 3 ... devices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "56812019-b1b9-4896-816e-7278de35a502",
   "metadata": {},
   "outputs": [],
   "source": [
    "n_users_per_ndev = n_devices_per_user.groupby('n_devices', as_index=False).count().sort_values(\n",
    "    'user_id', ascending = False).rename(columns = {'user_id':'n_users'})\n",
    "n_users_per_ndev.reset_index(drop = True, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "3a591863-9db1-47e5-8036-5e96bbab7064",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>n_devices</th>\n",
       "      <th>n_users</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>5954</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>5853</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>5532</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   n_devices  n_users\n",
       "0          3     5954\n",
       "1          2     5853\n",
       "2          4     5532"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n_users_per_ndev.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "0f8256b9-9f65-4345-95ce-2708d717576a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (12,3))\n",
    "sns.barplot(data = n_users_per_ndev.query('n_devices < 50'), x = 'n_devices', y = 'n_users')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "10581f2b-73e8-46f6-8027-d3916d290b79",
   "metadata": {},
   "source": [
    "**Findings:**  \n",
    "a) Usage of 3 devices and then 2 devices are the most frequent cases.  \n",
    "b) Then, the distribution of n_users by number of devices declines smoothly.\n",
    "\n",
    "**Thoughts & Question:**  \n",
    "I can assume what reasons (different gadgets, new ones, trading from corporate computer ...) are behind the number of devices used by a user from 2 to 5 and even up to 5-10 devices but by now it's not clear what reason is behind usage of 10+ devices by a user except the cases mentioned above ... "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e6dc722e-171c-4488-9b97-33f0d85a2bee",
   "metadata": {},
   "source": [
    "#### Do we have different users using same devices?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "11fdc41c-6b99-4782-83ac-4b4b14bb4ebc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>device_id</th>\n",
       "      <th>n_users_same_dev</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3cced232-d792-40e3-bdc7-ddff4ac02693</td>\n",
       "      <td>133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>08000000-0000-0000-0000-000000000000</td>\n",
       "      <td>121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>PtX2Y9zTfhULD7QHCr07</td>\n",
       "      <td>68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3BC2D0F6-0DFE-5A64-BF98-31810536B53E</td>\n",
       "      <td>67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>66210d5f-a85d-4e60-bae8-cb68de27029f</td>\n",
       "      <td>52</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              device_id  n_users_same_dev\n",
       "0  3cced232-d792-40e3-bdc7-ddff4ac02693               133\n",
       "1  08000000-0000-0000-0000-000000000000               121\n",
       "2                  PtX2Y9zTfhULD7QHCr07                68\n",
       "3  3BC2D0F6-0DFE-5A64-BF98-31810536B53E                67\n",
       "4  66210d5f-a85d-4e60-bae8-cb68de27029f                52"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_same_dev = df_ids.groupby('device_id', as_index=False).count().sort_values('user_id', ascending = False).reset_index(drop = True)\n",
    "df_same_dev.rename(columns = {'user_id':'n_users_same_dev'}, inplace = True)\n",
    "df_same_dev.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "31c61872-b086-4099-96cf-cdc3f9f0ca1a",
   "metadata": {},
   "source": [
    "This is strange that many different users use same devices and there are cases with a lot of user id(s) per one device ... that can be sign of fraud.  \n",
    "\n",
    "Let's mark users using same devices together with number of devices per user at the final dataframe to be used for ML analysis "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "8ac92e99-29aa-4bc0-ae02-25af981f4e59",
   "metadata": {},
   "outputs": [],
   "source": [
    "temp = df_ids.merge(df_same_dev, how = 'left', on = 'device_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "48d9af2e-9575-4666-a5ea-79cd182a6705",
   "metadata": {},
   "outputs": [],
   "source": [
    "temp = temp.groupby('user_id', as_index=False).agg({'device_id':'count', 'n_users_same_dev':'max'}) # leaving max number of used same devices\n",
    "temp.rename(columns = {'device_id':'n_devices'}, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "5d1aef89-8556-49b0-b39f-e51053f006e7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>n_devices</th>\n",
       "      <th>n_users_same_dev</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2352922</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2850718</td>\n",
       "      <td>149</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3283384</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3726424</td>\n",
       "      <td>64</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3994731</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id  n_devices  n_users_same_dev\n",
       "0  2352922          9                 1\n",
       "1  2850718        149                 1\n",
       "2  3283384         11                 1\n",
       "3  3726424         64                 1\n",
       "4  3994731          1                 6"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "8df06166-7923-4669-9c36-bae9ba662c43",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "temp['n_users_same_dev'].plot(kind = 'hist', bins = 50)\n",
    "plt.yscale('log')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ef3e26d-4416-41d8-bc76-143b63d29886",
   "metadata": {},
   "source": [
    "#### At temp dataframe we can see number of devices used by each user_id and number of user_id(s) using same device_id. For example, user_id = 3994731 use the same device which is used by another 5 user_id(s), so, totally, same device is used by 6 different accounts."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "efd9f9d4-cf5d-4e2f-b7cf-2ea8f8351015",
   "metadata": {},
   "source": [
    "#### Let's analyze the distribution of trades by hour"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "f64b35dd-b801-47d2-9d4a-3a17fcc96fe5",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_trade['hour'] = df_trade['create_at'].dt.hour\n",
    "counts = df_trade.groupby('hour')['id'].count()\n",
    "plt.bar(counts.index, counts.values)\n",
    "plt.xlabel('hour')\n",
    "plt.ylabel('number of trades')\n",
    "plt.title('Trades distribution by hour')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "581edc67-84c5-4fb7-a552-14cfc6733196",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_trade['hour_close'] = df_trade['close_at'].dt.hour\n",
    "counts = df_trade.groupby('hour_close')['id'].count()\n",
    "plt.bar(counts.index, counts.values)\n",
    "plt.xlabel('hour_close')\n",
    "plt.ylabel('number of trades')\n",
    "plt.title('Trades distribution by close hour')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a2a228d-24ff-467c-98bc-fc41616f3248",
   "metadata": {},
   "source": [
    "#### **Findings:**  \n",
    "a) There is no visual difference between create_at and close_at as expected as we deal with short-term options,  \n",
    "b) The most active hours for trading are from 12 to 20 hours (actually, from 12:00 till 20:59),  \n",
    "c) Next activity level can be observed at the periods from 0:00 till 2:59 (df.hours from 0 to 2), from 9:00 till 11:59, from 21:00 till 23:59  \n",
    "d) Almost \"silence\" period is from 3:00 till 8:59 (df.hours from 3 to 8),  \n",
    "\n",
    "Further I will add the column with the characteristics for every trade showing above levels of trade activity: 1 for high, 2 for middle, 3 for low for fraud analysis as some fraud trades can be executed when the activity is low and low attention from supervisors exists. Afterhand I will count for each user how many trades were executed at high-level period, middle-level and low-level-periods."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "71cf4f00-6178-44d2-bbbc-ed4768590bb4",
   "metadata": {},
   "outputs": [],
   "source": [
    "high_act = [12, 13, 14, 15, 16, 17, 18, 19, 20]\n",
    "middle_act = [0, 1, 2, 9, 10, 11, 21, 22, 23]\n",
    "low_act = [3, 4, 5, 6, 7, 8]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "ecbfca6c-7e19-48ff-9c7b-e219f43d04eb",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_trade['act_level'] = df_trade['hour'].apply(lambda x: 1 if x in high_act else 2 if x in middle_act else 3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27b63599-74b8-4c58-a80a-a2fd0c83da28",
   "metadata": {},
   "source": [
    "#### Let's have a look at pnl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "a961dd98-f568-43b6-9ace-b8315d7f82cd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_trade['pnl_total_enrolled'].plot(kind = 'hist', bins = 50, logy = True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14ffc3c7-e10d-4d5c-94a9-488e6a231908",
   "metadata": {},
   "source": [
    "#### Overall, it is not normal distribution but reminds it ... can we mark those users which have some outstaning profits and crucial losses?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "fc8b1c84-e952-4eb2-9e90-08382a9602fe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    907637.000000\n",
       "mean          6.974901\n",
       "std          28.154363\n",
       "min           0.000002\n",
       "25%           0.473175\n",
       "50%           1.000000\n",
       "75%           3.623520\n",
       "max        1544.560000\n",
       "Name: pnl_total_enrolled, dtype: float64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_trade.query('pnl_total_enrolled>0')['pnl_total_enrolled'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9789e56e-22eb-4337-8a29-e42b430f3a2e",
   "metadata": {},
   "source": [
    "#### I will use approximately 2 std added to mean value (~ +60/-60) to mark questionable trades with high or low pnl(s)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "9acb307d-3078-4ecd-9f7d-034823094b47",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "23141"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_trade.query('pnl_total_enrolled>60')['pnl_total_enrolled'].count() # only 23141 vs 1700000+ trades"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "865410b8-1027-43a9-ac95-26956f27ed26",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_trade['quest_pnl'] = df_trade['pnl_total_enrolled'].apply(lambda x: 1 if x>60 else 1 if x<-60 else 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dfc80ab7-7f59-419a-b61c-af8327706f8e",
   "metadata": {},
   "source": [
    "#### Let's see max and min pnl(s) at different hours"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "48f39bc9-0b98-47ae-99c7-7c9c16221f00",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "hour   0:\tmax pnl:     826.22\t min pnl: -1144.76\n",
      "hour   1:\tmax pnl:     955.75\t min pnl: -902.35\n",
      "hour   2:\tmax pnl:  569.10253\t min pnl: -522.768675\n",
      "hour   3:\tmax pnl: 983.408675\t min pnl: -1140.0\n",
      "hour   4:\tmax pnl: 557.061968\t min pnl: -813.225034\n",
      "hour   5:\tmax pnl:     167.95\t min pnl: -197.58\n",
      "hour   6:\tmax pnl:      269.9\t min pnl: -359.85\n",
      "hour   7:\tmax pnl:   972.7541\t min pnl: -1185.462\n",
      "hour   8:\tmax pnl: 321.037399\t min pnl: -344.0055\n",
      "hour   9:\tmax pnl:      969.0\t min pnl: -1065.15\n",
      "hour  10:\tmax pnl: 733.751951\t min pnl: -786.383863\n",
      "hour  11:\tmax pnl: 639.151162\t min pnl: -933.549267\n",
      "hour  12:\tmax pnl:   782.7116\t min pnl: -688.526445\n",
      "hour  13:\tmax pnl:   776.8937\t min pnl: -810.01\n",
      "hour  14:\tmax pnl: 815.655882\t min pnl: -1091.37\n",
      "hour  15:\tmax pnl:     969.26\t min pnl: -1000.0\n",
      "hour  16:\tmax pnl:     1038.0\t min pnl: -1000.0\n",
      "hour  17:\tmax pnl:    1014.86\t min pnl: -1178.12\n",
      "hour  18:\tmax pnl:    1079.63\t min pnl: -1199.51\n",
      "hour  19:\tmax pnl:    1544.56\t min pnl: -1558.32\n",
      "hour  20:\tmax pnl:     1151.7\t min pnl: -1199.89\n",
      "hour  21:\tmax pnl:    1009.15\t min pnl: -593.62\n",
      "hour  22:\tmax pnl:    1009.15\t min pnl: -1187.23\n",
      "hour  23:\tmax pnl:     864.21\t min pnl: -1110.21\n"
     ]
    }
   ],
   "source": [
    "for i in list(df_trade['hour'].sort_values().unique()):\n",
    "    print(f\"hour {i:3}:\\tmax pnl: {df_trade.query('hour == @i')['pnl_total_enrolled'].max():10}\\t min pnl: {df_trade.query('hour == @i')['pnl_total_enrolled'].min()}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1707d63-074f-4e34-ab89-864576f61ce2",
   "metadata": {},
   "source": [
    "#### As there are significant pnl(s) at \"silence\" period we will mark also those trades"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "440d749a-af5c-4732-b9a8-e1c2b34c9412",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_trade['night_quest_pnl'] = df_trade.query('hour in @low_act')['pnl_total_enrolled'].apply(lambda x: 1 if x>60 else 1 if x<-60 else 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "79561c26-9b30-4d52-b590-6b6cf33e077b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "id                       \tNumber of unique items:1764494\t % of total rows: 100.0%\n",
      "user_id                  \tNumber of unique items:  81726\t % of total rows:   4.6%\n",
      "registration_date        \tNumber of unique items:  81726\t % of total rows:   4.6%\n",
      "fin_instrument           \tNumber of unique items:      5\t % of total rows:   0.0%\n",
      "instrument_type          \tNumber of unique items:      2\t % of total rows:   0.0%\n",
      "instrument_direction     \tNumber of unique items:      2\t % of total rows:   0.0%\n",
      "create_at                \tNumber of unique items:1736398\t % of total rows:  98.4%\n",
      "close_at                 \tNumber of unique items: 104679\t % of total rows:   5.9%\n",
      "instrument_expiration    \tNumber of unique items:  17980\t % of total rows:   1.0%\n",
      "close_reason             \tNumber of unique items:      3\t % of total rows:   0.0%\n",
      "platform                 \tNumber of unique items:     20\t % of total rows:   0.0%\n",
      "open_quote               \tNumber of unique items: 577557\t % of total rows:  32.7%\n",
      "close_quote              \tNumber of unique items:  85933\t % of total rows:   4.9%\n",
      "instrument_strike        \tNumber of unique items: 578642\t % of total rows:  32.8%\n",
      "initial_investment_enrolled\tNumber of unique items: 214957\t % of total rows:  12.2%\n",
      "pnl_total_enrolled       \tNumber of unique items: 529599\t % of total rows:  30.0%\n",
      "hour                     \tNumber of unique items:     24\t % of total rows:   0.0%\n",
      "hour_close               \tNumber of unique items:     24\t % of total rows:   0.0%\n",
      "act_level                \tNumber of unique items:      3\t % of total rows:   0.0%\n",
      "quest_pnl                \tNumber of unique items:      2\t % of total rows:   0.0%\n",
      "night_quest_pnl          \tNumber of unique items:      2\t % of total rows:   0.0%\n"
     ]
    }
   ],
   "source": [
    "# just to see some statistics\n",
    "for i in list(df_trade.columns):\n",
    "    print(f\"{i:25}\\tNumber of unique items:{df_trade[i].nunique():7}\\t % of total rows:{round(100*df_trade[i].nunique()/df_trade.shape[0],1):6}%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b10eef7-3a67-45c3-920c-3b8c16556eb5",
   "metadata": {},
   "source": [
    "#### Let's have a look at categorical features:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "71addedd-1815-4570-a297-b7931c26df89",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(nrows = 2, ncols = 2, figsize=(8,5))\n",
    "sns.countplot(data = df_trade, x = 'fin_instrument', ax = axes[0,0])\n",
    "axes[0,0].set_title('Number of rows by a fin_instrument')\n",
    "sns.countplot(data = df_trade, x = 'instrument_type', ax = axes[0,1])\n",
    "axes[0,1].set_title('Number of rows by instrument type')\n",
    "sns.countplot(data = df_trade, x = 'instrument_direction', ax = axes[1,0])\n",
    "axes[1,0].set_title('Number of rows by instrument_direction type')\n",
    "sns.countplot(data = df_trade, x = 'close_reason', ax = axes[1,1])\n",
    "axes[1,1].set_title('Number of rows by close_reason type')\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e4a917c0-8373-463b-8e3a-23eb285379ce",
   "metadata": {},
   "source": [
    "**Findings:**  \n",
    "a) The biggest number of trades was executed with XRPUSD.  \n",
    "b) Digital options were more popular vs. turbo options.  \n",
    "c) Numbers of call and put options were almost the same."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "c90bdcf6-7e5f-4071-b8c9-0ff17bd7e759",
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "counter = Counter(df_trade.platform)\n",
    "order = [value[0] for value in counter.most_common()]#[::-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "cb46c938-db5e-4d05-9d35-e40b400448e1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (12,3))\n",
    "sns.countplot(data = df_trade, x = 'platform', order = order, palette = 'plasma')\n",
    "plt.xticks(rotation = 90)\n",
    "plt.title('Platforms split')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b621188-e9ed-475d-9ed3-dc36c34cf71e",
   "metadata": {},
   "source": [
    "**Findings:**  \n",
    "a) Top 3 platforms are WinGL, Android, WebGL.  \n",
    "b) Unknown platform is in 4th place."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "901623eb-ef43-4247-aaaf-ad856132826c",
   "metadata": {},
   "source": [
    "### Let's merge temp dataframe with number of devices and numbers of user_id(s) using same devices with df_trade and investigate new dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "6a3d9d47-5d01-4791-8808-bd57734e1bd2",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df_trade.merge(temp, how = 'left', on = 'user_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "957f4e29-a216-41e9-a03a-e58d9fe3a96a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>user_id</th>\n",
       "      <th>registration_date</th>\n",
       "      <th>fin_instrument</th>\n",
       "      <th>instrument_type</th>\n",
       "      <th>instrument_direction</th>\n",
       "      <th>create_at</th>\n",
       "      <th>close_at</th>\n",
       "      <th>instrument_expiration</th>\n",
       "      <th>close_reason</th>\n",
       "      <th>...</th>\n",
       "      <th>instrument_strike</th>\n",
       "      <th>initial_investment_enrolled</th>\n",
       "      <th>pnl_total_enrolled</th>\n",
       "      <th>hour</th>\n",
       "      <th>hour_close</th>\n",
       "      <th>act_level</th>\n",
       "      <th>quest_pnl</th>\n",
       "      <th>night_quest_pnl</th>\n",
       "      <th>n_devices</th>\n",
       "      <th>n_users_same_dev</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9653594472</td>\n",
       "      <td>585118157</td>\n",
       "      <td>2019-04-05 17:12:46.000000</td>\n",
       "      <td>XRPUSD</td>\n",
       "      <td>turbo-option</td>\n",
       "      <td>call</td>\n",
       "      <td>2022-06-17 19:41:03.216000+00:00</td>\n",
       "      <td>2022-06-17 19:42:00+00:00</td>\n",
       "      <td>2022-06-17 19:42:00+00:00</td>\n",
       "      <td>expired</td>\n",
       "      <td>...</td>\n",
       "      <td>0.314705</td>\n",
       "      <td>22.610000</td>\n",
       "      <td>19.210000</td>\n",
       "      <td>19</td>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21.0</td>\n",
       "      <td>47.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9641975390</td>\n",
       "      <td>1456695825</td>\n",
       "      <td>2021-08-12 02:28:57.928101</td>\n",
       "      <td>XRPUSD</td>\n",
       "      <td>turbo-option</td>\n",
       "      <td>call</td>\n",
       "      <td>2022-06-14 00:23:15.137000+00:00</td>\n",
       "      <td>2022-06-14 00:24:00+00:00</td>\n",
       "      <td>2022-06-14 00:24:00+00:00</td>\n",
       "      <td>expired</td>\n",
       "      <td>...</td>\n",
       "      <td>0.308125</td>\n",
       "      <td>0.390000</td>\n",
       "      <td>-0.390000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>15.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15634194734</td>\n",
       "      <td>220188644</td>\n",
       "      <td>2017-03-25 01:22:50.000000</td>\n",
       "      <td>XRPUSD</td>\n",
       "      <td>digital-option</td>\n",
       "      <td>call</td>\n",
       "      <td>2022-06-13 20:18:13.091000+00:00</td>\n",
       "      <td>2022-06-13 20:19:00+00:00</td>\n",
       "      <td>2022-06-13 20:19:00+00:00</td>\n",
       "      <td>expired</td>\n",
       "      <td>...</td>\n",
       "      <td>0.323725</td>\n",
       "      <td>0.391288</td>\n",
       "      <td>-0.391288</td>\n",
       "      <td>20</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>351.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            id     user_id          registration_date fin_instrument  \\\n",
       "0   9653594472   585118157 2019-04-05 17:12:46.000000         XRPUSD   \n",
       "1   9641975390  1456695825 2021-08-12 02:28:57.928101         XRPUSD   \n",
       "2  15634194734   220188644 2017-03-25 01:22:50.000000         XRPUSD   \n",
       "\n",
       "  instrument_type instrument_direction                        create_at  \\\n",
       "0    turbo-option                 call 2022-06-17 19:41:03.216000+00:00   \n",
       "1    turbo-option                 call 2022-06-14 00:23:15.137000+00:00   \n",
       "2  digital-option                 call 2022-06-13 20:18:13.091000+00:00   \n",
       "\n",
       "                   close_at     instrument_expiration close_reason  ...  \\\n",
       "0 2022-06-17 19:42:00+00:00 2022-06-17 19:42:00+00:00      expired  ...   \n",
       "1 2022-06-14 00:24:00+00:00 2022-06-14 00:24:00+00:00      expired  ...   \n",
       "2 2022-06-13 20:19:00+00:00 2022-06-13 20:19:00+00:00      expired  ...   \n",
       "\n",
       "  instrument_strike  initial_investment_enrolled  pnl_total_enrolled  hour  \\\n",
       "0          0.314705                    22.610000           19.210000    19   \n",
       "1          0.308125                     0.390000           -0.390000     0   \n",
       "2          0.323725                     0.391288           -0.391288    20   \n",
       "\n",
       "   hour_close  act_level  quest_pnl  night_quest_pnl  n_devices  \\\n",
       "0          19          1          0              NaN       21.0   \n",
       "1           0          2          0              NaN       15.0   \n",
       "2          20          1          0              NaN      351.0   \n",
       "\n",
       "   n_users_same_dev  \n",
       "0              47.0  \n",
       "1               3.0  \n",
       "2               1.0  \n",
       "\n",
       "[3 rows x 23 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "c4497668-caa9-4ce3-b11d-416d6d773b68",
   "metadata": {},
   "outputs": [],
   "source": [
    "# let's replace all NaN in the columns: n_devices and n_users_same_dev by 0s as we don't have the info about 720 user ids.\n",
    "df['n_devices'].fillna(0, inplace=True)\n",
    "df['n_users_same_dev'].fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c6c7ce23-a2bc-42c0-a135-e75d31291a02",
   "metadata": {},
   "source": [
    "#### Let's investigate pnl and number of wins (profitable trades) vs losses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "5f82dc99-b9fb-4d85-a2e6-2113b7368647",
   "metadata": {},
   "outputs": [],
   "source": [
    "temp = df[['user_id','pnl_total_enrolled']].sort_values('pnl_total_enrolled', ascending = False)\n",
    "temp['win'] = temp['pnl_total_enrolled'].apply(lambda x: 1 if x>0 else -1) # assign 1 for positive trade and -1 for negative trade\n",
    "temp = temp.groupby('user_id', as_index = False).agg({'pnl_total_enrolled':'sum', 'win':'sum'}).rename(columns = {'pnl_total_enrolled':'pnl_total', 'win':'n_wins'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "f964eae0-ea81-4e0c-8fee-f805bc161eec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pnl_total</th>\n",
       "      <th>n_wins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>81726.000000</td>\n",
       "      <td>81726.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>12.957315</td>\n",
       "      <td>0.621344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>359.739658</td>\n",
       "      <td>23.896946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-3215.437469</td>\n",
       "      <td>-4972.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-2.701822</td>\n",
       "      <td>-2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-0.374824</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.640278</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>25270.761223</td>\n",
       "      <td>1402.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          pnl_total        n_wins\n",
       "count  81726.000000  81726.000000\n",
       "mean      12.957315      0.621344\n",
       "std      359.739658     23.896946\n",
       "min    -3215.437469  -4972.000000\n",
       "25%       -2.701822     -2.000000\n",
       "50%       -0.374824      0.000000\n",
       "75%        1.640278      2.000000\n",
       "max    25270.761223   1402.000000"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp[['pnl_total','n_wins']].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "4b821d74-24c1-4e39-8139-7c335211fc4e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot(data = temp, x = 'user_id', y = 'n_wins')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7cc538a7-9e00-4c40-9130-ca54910247a3",
   "metadata": {},
   "source": [
    "#### **Findings:**  \n",
    "a) there are clear outliers with number of wins and total pnl above 75% percentile.  \n",
    "b) The most of users in terms of number of wins (sum of 1s and -1s) are between -2 and 2.  \n",
    "b) It's interesting that 75% percentile of pnl_total is less than 25% percentile in absolute value that may say that there are some users with very high earnings."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe9c0070-266a-4414-93c5-35a79de6591f",
   "metadata": {},
   "source": [
    "## 2. ML Modeling: adding new features."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "79a6f1d0-6aa4-4d87-a75d-57f426c14e36",
   "metadata": {},
   "source": [
    "#### Let's add as extra column numbers of trades by a user."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "92df32e8-502e-451e-a247-56eb0c02ee29",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.merge(df.groupby('user_id', as_index = False).agg({'id':'count'}).\\\n",
    "        sort_values('id', ascending = False).rename(columns = {'id':'n_trades'}), how = 'left', on = 'user_id')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e89289c-f934-4b59-a5a4-4370cec8084d",
   "metadata": {},
   "source": [
    "#### Let's add count of used calls and puts by a user as extra column and replace NaNs by 0s."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "54a5d893-accc-44b1-94c0-f9bb68e6d9f6",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.merge(df.query('instrument_direction == \"put\"').groupby('user_id', as_index = False)['id'].count().\\\n",
    "        rename(columns = {'id':'n_put'}), how = 'left', on = 'user_id')\n",
    "df = df.merge(df.query('instrument_direction == \"call\"').groupby('user_id', as_index = False)['id'].count().\\\n",
    "        rename(columns = {'id':'n_call'}), how = 'left', on = 'user_id')\n",
    "df['n_put'].fillna(0, inplace=True)\n",
    "df['n_call'].fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a00277ad-73a6-4ade-9bb7-b2ffbe4a169c",
   "metadata": {},
   "source": [
    "#### Let's add count of used instrument types by a user as extra column and replace NaNs by 0s."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "a2fc331b-a094-4574-a783-e95c9216b9d1",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.merge(df.query('instrument_type == \"turbo-option\"').groupby('user_id', as_index = False)['id'].count().\\\n",
    "        rename(columns = {'id':'n_turbo'}),  how = 'left', on = 'user_id')\n",
    "df = df.merge(df.query('instrument_type == \"digital-option\"').groupby('user_id', as_index = False)['id'].count().\\\n",
    "        rename(columns = {'id':'n_digital'}),  how = 'left', on = 'user_id')\n",
    "df['n_turbo'].fillna(0, inplace=True)\n",
    "df['n_digital'].fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8677170e-ad22-4167-a136-a30678112487",
   "metadata": {},
   "source": [
    "#### Let's add count of used platforms by a user as extra column and replace NaNs by 0s."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "c134e4b2-50c6-4f08-8e19-b02e4f17d1ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "ls_pl = list(df.platform.unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "0d7aafe1-5c02-4927-b41a-bc3b3951bc8f",
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in ls_pl:\n",
    "    df = df.merge(df.query('platform == @i').groupby('user_id', as_index = False)['id'].count().\\\n",
    "        rename(columns = {'id':'n_'+i}), how = 'left', on = 'user_id')\n",
    "    df['n_'+i].fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2d14611-bafb-4894-9cc1-57e008270381",
   "metadata": {},
   "source": [
    "#### Let's add pnl_total and count of wins vs losses by a user from the temp dataframe above as extra column and replace NaNs by 0s."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "147c8dad-2d65-4479-89d7-aaef02b7dcb0",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.merge(temp, how = 'left', on = 'user_id')\n",
    "df['pnl_total'].fillna(0, inplace=True)\n",
    "df['n_wins'].fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "da8858d5-3582-498b-92d2-614b9acce537",
   "metadata": {},
   "source": [
    "#### Let's add total investment by a user as extra column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "4d621b9a-3e42-412e-9ecd-b6f2d065547c",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.merge(df.groupby('user_id', as_index = False).agg({'initial_investment_enrolled':'sum'}).\\\n",
    "sort_values('initial_investment_enrolled', ascending = False).rename(columns = {'initial_investment_enrolled':'total_investment'}),\\\n",
    "              how = 'left', on = 'user_id')\n",
    "df['total_investment'].fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "de3d6f63-c4a4-4382-a89a-6bdb18652e73",
   "metadata": {},
   "source": [
    "#### Let's calculate the number of trade as per Activity level by a user as extra column and replace NaNs by 0s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "122c6d2c-6394-4e77-9ff6-924f16460804",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.merge(df.groupby('user_id', as_index = False).agg({'act_level':'mean'}).\\\n",
    "        rename(columns = {'act_level':'avg_act_level'}), how = 'left', on = 'user_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "47764b44-b372-4984-a7e1-015f5ff18b49",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['avg_act_level'].fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c983e3bb-01ad-4f32-bffd-f197f06c4c62",
   "metadata": {},
   "source": [
    "#### Let's add also a year of registration ... may be, there is some dependency on it like fraud customers joined recently"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "9de92264-ca0c-4673-a768-e348b67ec441",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['reg_year'] = df['registration_date'].dt.year"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e440f1c9-3027-416e-a7b2-69cff5ab8525",
   "metadata": {},
   "source": [
    "#### Let's calculate the number of questionable trades with either high pnl or with high pnl at \"silence: period by a user as extra column and replace NaNs by 0s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "373dec01-4043-466b-bc46-8948fca68485",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.merge(df.groupby('user_id', as_index = False).agg({'quest_pnl':'sum', 'night_quest_pnl':'sum'}).\\\n",
    "        rename(columns = {'quest_pnl':'n_quest_pnl', 'night_quest_pnl':'n_night_quest_pnl'}), how = 'left', on = 'user_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "d2b1d3d2-41ba-4b74-8440-308f7d0583e3",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['n_quest_pnl'].fillna(0, inplace=True)\n",
    "df['n_night_quest_pnl'].fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "24a9632f-07fe-49e4-a4e2-6f0d06ec913c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['id', 'user_id', 'registration_date', 'fin_instrument',\n",
       "       'instrument_type', 'instrument_direction', 'create_at', 'close_at',\n",
       "       'instrument_expiration', 'close_reason', 'platform', 'open_quote',\n",
       "       'close_quote', 'instrument_strike', 'initial_investment_enrolled',\n",
       "       'pnl_total_enrolled', 'hour', 'hour_close', 'act_level', 'quest_pnl',\n",
       "       'night_quest_pnl', 'n_devices', 'n_users_same_dev', 'n_trades', 'n_put',\n",
       "       'n_call', 'n_turbo', 'n_digital', 'n_Unknown platform', 'n_WebGL',\n",
       "       'n_WinGL', 'n_Desktop Mobile', 'n_Android', 'n_IQ Option X',\n",
       "       'n_WEB_MOBILE', 'n_Ios', 'n_Android pwa for iqoption', 'n_MacGL',\n",
       "       'n_Ios CFD', 'n_Android Facebook', 'n_IQ Option Wallet Web',\n",
       "       'n_Android Vertical', 'n_CapitalBear.Android', 'n_IOS pwa for iqoption',\n",
       "       'n_CapitalBear.WebGL', 'n_CapitalBear.WinGL', 'n_Exnova.WebGL',\n",
       "       'n_CapitalBear.AndroidX', 'pnl_total', 'n_wins', 'total_investment',\n",
       "       'avg_act_level', 'reg_year', 'n_quest_pnl', 'n_night_quest_pnl'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c5f4c4f6-cd1a-4fa8-a714-f525d6453b2f",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "5008ef75-acc2-40eb-b13c-727745570d85",
   "metadata": {},
   "source": [
    "#### Preliminary dataframe with user_id and only needed features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "3b230441-fde0-43ef-ad6f-c99ab9a71167",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[['user_id', 'n_devices', 'n_users_same_dev', 'n_trades', 'n_put',\n",
    "       'n_call', 'n_turbo', 'n_digital', 'n_Unknown platform', 'n_WebGL',\n",
    "       'n_WinGL', 'n_Desktop Mobile', 'n_Android', 'n_IQ Option X',\n",
    "       'n_WEB_MOBILE', 'n_Ios', 'n_Android pwa for iqoption', 'n_MacGL',\n",
    "       'n_Ios CFD', 'n_Android Facebook', 'n_IQ Option Wallet Web',\n",
    "       'n_Android Vertical', 'n_CapitalBear.Android', 'n_IOS pwa for iqoption',\n",
    "       'n_CapitalBear.WebGL', 'n_CapitalBear.WinGL', 'n_Exnova.WebGL',\n",
    "       'n_CapitalBear.AndroidX', 'pnl_total', 'n_wins', 'total_investment',\n",
    "       'avg_act_level', 'reg_year', 'n_quest_pnl', 'n_night_quest_pnl']]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08ca6512-36f7-4bde-91a8-333db0de218a",
   "metadata": {},
   "source": [
    "#### We need to drop duplicates as we added features for every repeating row with user_id"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "70de5952-b01d-406f-b733-7c2963c4733a",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop_duplicates(inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "2af3f73c-29d3-44e3-b5cb-3d41f0a0c043",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Index: 81726 entries, 0 to 1764481\n",
      "Data columns (total 35 columns):\n",
      " #   Column                      Non-Null Count  Dtype  \n",
      "---  ------                      --------------  -----  \n",
      " 0   user_id                     81726 non-null  int64  \n",
      " 1   n_devices                   81726 non-null  float64\n",
      " 2   n_users_same_dev            81726 non-null  float64\n",
      " 3   n_trades                    81726 non-null  int64  \n",
      " 4   n_put                       81726 non-null  float64\n",
      " 5   n_call                      81726 non-null  float64\n",
      " 6   n_turbo                     81726 non-null  float64\n",
      " 7   n_digital                   81726 non-null  float64\n",
      " 8   n_Unknown platform          81726 non-null  float64\n",
      " 9   n_WebGL                     81726 non-null  float64\n",
      " 10  n_WinGL                     81726 non-null  float64\n",
      " 11  n_Desktop Mobile            81726 non-null  float64\n",
      " 12  n_Android                   81726 non-null  float64\n",
      " 13  n_IQ Option X               81726 non-null  float64\n",
      " 14  n_WEB_MOBILE                81726 non-null  float64\n",
      " 15  n_Ios                       81726 non-null  float64\n",
      " 16  n_Android pwa for iqoption  81726 non-null  float64\n",
      " 17  n_MacGL                     81726 non-null  float64\n",
      " 18  n_Ios CFD                   81726 non-null  float64\n",
      " 19  n_Android Facebook          81726 non-null  float64\n",
      " 20  n_IQ Option Wallet Web      81726 non-null  float64\n",
      " 21  n_Android Vertical          81726 non-null  float64\n",
      " 22  n_CapitalBear.Android       81726 non-null  float64\n",
      " 23  n_IOS pwa for iqoption      81726 non-null  float64\n",
      " 24  n_CapitalBear.WebGL         81726 non-null  float64\n",
      " 25  n_CapitalBear.WinGL         81726 non-null  float64\n",
      " 26  n_Exnova.WebGL              81726 non-null  float64\n",
      " 27  n_CapitalBear.AndroidX      81726 non-null  float64\n",
      " 28  pnl_total                   81726 non-null  float64\n",
      " 29  n_wins                      81726 non-null  int64  \n",
      " 30  total_investment            81726 non-null  float64\n",
      " 31  avg_act_level               81726 non-null  float64\n",
      " 32  reg_year                    81726 non-null  int32  \n",
      " 33  n_quest_pnl                 81726 non-null  int64  \n",
      " 34  n_night_quest_pnl           81726 non-null  float64\n",
      "dtypes: float64(30), int32(1), int64(4)\n",
      "memory usage: 22.1 MB\n"
     ]
    }
   ],
   "source": [
    "df.info() # no NaNs and no objects. Number of rows is equal to the number of unique user_id(s)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "c70adabe-eee8-41e7-b936-55446760db45",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.reset_index(drop = True, inplace = True) # to fix the mess with indexes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "c865a7b4-3e39-45d1-916c-028762eaee6c",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.drop('user_id', axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "c33ca55b-e861-49fb-9b48-65c20fb08060",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "01ffdc8b-fa68-4e56-bb4a-c688d56e5858",
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = MinMaxScaler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "f690ef2f-5d89-46de-9ed1-8db233f894be",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_scaled = scaler.fit_transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "19de912b-e02d-4162-ae39-56cb91abc4fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.cluster import KMeans"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "307ac17f-f32c-4367-8852-6571ecc41748",
   "metadata": {
    "id": "iWfhUMk1o3f6"
   },
   "source": [
    "KMeans will minimize the sum of squared distances between objects of clusters and their centroids, the value is found in the inertia_. Let's calculate inertia values for different number of clusters. It should eager to some minimal value, ideally, to 0)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "cd19b322-bf8e-461b-aec0-12ecf453d5f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "k_inertia = []\n",
    "ks = range(1,50)\n",
    "\n",
    "for k in ks:\n",
    "    clf_kmeans = KMeans(n_clusters=k, random_state = 101)\n",
    "    clusters_kmeans = clf_kmeans.fit_predict(X_scaled, )\n",
    "    k_inertia.append(clf_kmeans.inertia_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "88175401-4edd-4921-979f-406df07c3ade",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 554
    },
    "id": "2UlKITFqo3f8",
    "outputId": "5fc8f1c1-df30-42e1-a436-1f8fcbd9e30e"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(ks, k_inertia)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "0a40e004-8036-43e6-a349-e1d9532917bd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6784.879786</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3294.740603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2251.919343</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1910.064927</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1583.577270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1320.386365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1144.951507</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>916.738777</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>782.499999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>704.508962</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>635.657613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>584.344272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>540.623352</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>507.683743</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>475.032098</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>451.553227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>428.330868</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>410.128292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>393.879273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>379.723824</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>369.763528</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>353.966160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>332.488537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>324.751357</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>318.001800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>307.755262</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>294.814076</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>286.827455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>274.867733</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>273.085275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>254.993994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>255.646481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>246.435658</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>242.701885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>234.566773</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>228.828877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>225.971713</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>223.863137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>214.679170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>216.663413</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>209.181228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>205.009023</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>200.328169</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>195.148980</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>192.341487</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>189.256276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>186.433113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>184.330895</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>177.370262</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              0\n",
       "1   6784.879786\n",
       "2   3294.740603\n",
       "3   2251.919343\n",
       "4   1910.064927\n",
       "5   1583.577270\n",
       "6   1320.386365\n",
       "7   1144.951507\n",
       "8    916.738777\n",
       "9    782.499999\n",
       "10   704.508962\n",
       "11   635.657613\n",
       "12   584.344272\n",
       "13   540.623352\n",
       "14   507.683743\n",
       "15   475.032098\n",
       "16   451.553227\n",
       "17   428.330868\n",
       "18   410.128292\n",
       "19   393.879273\n",
       "20   379.723824\n",
       "21   369.763528\n",
       "22   353.966160\n",
       "23   332.488537\n",
       "24   324.751357\n",
       "25   318.001800\n",
       "26   307.755262\n",
       "27   294.814076\n",
       "28   286.827455\n",
       "29   274.867733\n",
       "30   273.085275\n",
       "31   254.993994\n",
       "32   255.646481\n",
       "33   246.435658\n",
       "34   242.701885\n",
       "35   234.566773\n",
       "36   228.828877\n",
       "37   225.971713\n",
       "38   223.863137\n",
       "39   214.679170\n",
       "40   216.663413\n",
       "41   209.181228\n",
       "42   205.009023\n",
       "43   200.328169\n",
       "44   195.148980\n",
       "45   192.341487\n",
       "46   189.256276\n",
       "47   186.433113\n",
       "48   184.330895\n",
       "49   177.370262"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(k_inertia, index = list(range(1, len(k_inertia)+1)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aa64d4a4-76e0-477f-b88e-35bb490af4bc",
   "metadata": {},
   "source": [
    "#### Let's choose optimal number of clusters taking into account inertia value and change vs. previous ones"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "e6183a46-7b44-4890-80ad-9490b3408369",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "diff = np.diff(k_inertia)\n",
    "plt.plot(ks[1:], diff)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "4effbae8-6eb2-4c4b-b599-5a417366351d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "diff_r = diff[1:] / diff[:-1]\n",
    "plt.plot(ks[1:-1], diff_r)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "id": "b7abafaf-d90f-4bee-8204-80d7d87d1c13",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.298791</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.327817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.955049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.806128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.666569</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1.300840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.588218</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.580987</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.882811</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.745277</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.852038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.753406</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.991258</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.719072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.989075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.783838</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.892677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.871157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.703637</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>1.586034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>1.359570</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>0.360244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>0.872354</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>1.518105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>1.262981</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>0.617147</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>1.497470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>0.149038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>10.149627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>-0.036066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>-14.116487</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>0.405368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>2.178791</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>0.705325</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>0.497946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>0.737996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>4.355530</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>-0.216055</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>-3.770801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>0.557619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>1.121914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>1.106462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>0.542072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>1.098921</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>0.915063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>0.744632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>3.311090</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            0\n",
       "1    0.298791\n",
       "2    0.327817\n",
       "3    0.955049\n",
       "4    0.806128\n",
       "5    0.666569\n",
       "6    1.300840\n",
       "7    0.588218\n",
       "8    0.580987\n",
       "9    0.882811\n",
       "10   0.745277\n",
       "11   0.852038\n",
       "12   0.753406\n",
       "13   0.991258\n",
       "14   0.719072\n",
       "15   0.989075\n",
       "16   0.783838\n",
       "17   0.892677\n",
       "18   0.871157\n",
       "19   0.703637\n",
       "20   1.586034\n",
       "21   1.359570\n",
       "22   0.360244\n",
       "23   0.872354\n",
       "24   1.518105\n",
       "25   1.262981\n",
       "26   0.617147\n",
       "27   1.497470\n",
       "28   0.149038\n",
       "29  10.149627\n",
       "30  -0.036066\n",
       "31 -14.116487\n",
       "32   0.405368\n",
       "33   2.178791\n",
       "34   0.705325\n",
       "35   0.497946\n",
       "36   0.737996\n",
       "37   4.355530\n",
       "38  -0.216055\n",
       "39  -3.770801\n",
       "40   0.557619\n",
       "41   1.121914\n",
       "42   1.106462\n",
       "43   0.542072\n",
       "44   1.098921\n",
       "45   0.915063\n",
       "46   0.744632\n",
       "47   3.311090"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(diff_r, index = list(range(1, len(diff_r)+1)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "451359d2-860c-4f95-8c72-2d87b00b23f4",
   "metadata": {},
   "source": [
    "#### Let's choose number of clusters equal to 22 as the changes vs previous values became unstable ... as a test we can choose afterwards, for example, number of clusters equal to higher number to compare the sets of users if we will find some suspicious cluster from fraud POV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "id": "9984199d-88ad-4085-91e3-74f1b64e5e07",
   "metadata": {},
   "outputs": [],
   "source": [
    "n_clusters = 22"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "id": "152359b9-8bf8-4862-aac7-61f7258cffbf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>KMeans(n_clusters=22, random_state=101)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" checked><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">KMeans</label><div class=\"sk-toggleable__content\"><pre>KMeans(n_clusters=22, random_state=101)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "KMeans(n_clusters=22, random_state=101)"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "kmeans = KMeans(n_clusters = n_clusters, random_state = 101)\n",
    "kmeans.fit(X_scaled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "id": "e2acff14-178e-4b0c-9349-9b1c732d8b19",
   "metadata": {},
   "outputs": [],
   "source": [
    "X['cluster'] = kmeans.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "id": "d542a6e3-6073-4c2a-83d4-a0313b72792f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cluster 0 has 4285 users\n",
      "Cluster 1 has 4311 users\n",
      "Cluster 2 has 18157 users\n",
      "Cluster 3 has 621 users\n",
      "Cluster 4 has 1285 users\n",
      "Cluster 5 has 254 users\n",
      "Cluster 6 has 2256 users\n",
      "Cluster 7 has 14901 users\n",
      "Cluster 8 has 12478 users\n",
      "Cluster 9 has 2138 users\n",
      "Cluster 10 has 921 users\n",
      "Cluster 11 has 2917 users\n",
      "Cluster 12 has 4738 users\n",
      "Cluster 13 has 73 users\n",
      "Cluster 14 has 491 users\n",
      "Cluster 15 has 920 users\n",
      "Cluster 16 has 2350 users\n",
      "Cluster 17 has 4538 users\n",
      "Cluster 18 has 1926 users\n",
      "Cluster 19 has 658 users\n",
      "Cluster 20 has 853 users\n",
      "Cluster 21 has 655 users\n"
     ]
    }
   ],
   "source": [
    "for i in list(X.cluster.sort_values().unique()):\n",
    "    print(f\"Cluster {i} has {X.query('cluster == @i').shape[0]} users\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "888d6b90-45ce-47d1-a8ab-39c42833b770",
   "metadata": {},
   "source": [
    "### Let's calulcate averages for every feature by a cluster and analyze the results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "id": "2d7e1fd4-7115-4a39-af47-3a50abb252c5",
   "metadata": {},
   "outputs": [],
   "source": [
    "result = pd.DataFrame(columns = X.columns)\n",
    "for i in range(0,n_clusters):\n",
    "    result = pd.concat([result, pd.DataFrame(X.query('cluster == @i').mean()).T], axis = 0)\n",
    "result.reset_index(drop = True, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "id": "538136d8-6d0b-4c0c-89de-e10bdfdc8062",
   "metadata": {},
   "outputs": [],
   "source": [
    "z = 'c'+str(n_clusters)+'.xlsx'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "id": "37007db5-4761-4586-90f3-f5f8d29b6553",
   "metadata": {},
   "outputs": [],
   "source": [
    "result.to_excel(z, index = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15d75439-504c-438f-b357-2d46514b490b",
   "metadata": {},
   "source": [
    "## **Conclusions:**  \n",
    "The most suspicious cluster is **cluster 13** with just 73 users. It appeared the \"champion\" in almost all important \"nominations\":  \n",
    "a) The highest number of devices per user,  \n",
    "b) Number of trades,  \n",
    "c) The highest number of used Unknown platforms and some others,  \n",
    "d) PnL was significantly higher than for other clusters together with number of wins as per the formula for the criteria,  \n",
    "e) Total investment is huge,  \n",
    "f) the most trades were executed in \"silence\" period (avg = 1.144)  \n",
    "g) number of high profits per a trade (all day and in \"silence\" period) was also higher that for other clusters \n",
    "\n",
    "I tested another n_clusters such as 25 and 40 for modeling and every time resulted clusters with similar output had around 65-75 users (the same users) that confirms the model with chosen features provides reasonable results.\n",
    "\n",
    "Excel file with color coding is attached."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "id": "2e2eb898-9f58-4cfe-be2b-e8627a22f736",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>13</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>n_devices</th>\n",
       "      <td>180.630137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_users_same_dev</th>\n",
       "      <td>23.013699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_trades</th>\n",
       "      <td>2210.671233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_put</th>\n",
       "      <td>1112.972603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_call</th>\n",
       "      <td>1097.698630</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_turbo</th>\n",
       "      <td>647.479452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_digital</th>\n",
       "      <td>1563.191781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_Unknown platform</th>\n",
       "      <td>555.301370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_WebGL</th>\n",
       "      <td>55.986301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_WinGL</th>\n",
       "      <td>1157.178082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_Desktop Mobile</th>\n",
       "      <td>0.178082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_Android</th>\n",
       "      <td>32.890411</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_IQ Option X</th>\n",
       "      <td>168.054795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_WEB_MOBILE</th>\n",
       "      <td>89.958904</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_Ios</th>\n",
       "      <td>150.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_Android pwa for iqoption</th>\n",
       "      <td>0.164384</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_MacGL</th>\n",
       "      <td>0.246575</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_Ios CFD</th>\n",
       "      <td>0.328767</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_Android Facebook</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_IQ Option Wallet Web</th>\n",
       "      <td>0.205479</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_Android Vertical</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_CapitalBear.Android</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_IOS pwa for iqoption</th>\n",
       "      <td>0.178082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_CapitalBear.WebGL</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_CapitalBear.WinGL</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_Exnova.WebGL</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_CapitalBear.AndroidX</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pnl_total</th>\n",
       "      <td>9344.598914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_wins</th>\n",
       "      <td>42.424658</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_investment</th>\n",
       "      <td>58429.725992</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avg_act_level</th>\n",
       "      <td>1.143996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reg_year</th>\n",
       "      <td>2020.260274</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_quest_pnl</th>\n",
       "      <td>281.452055</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_night_quest_pnl</th>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cluster</th>\n",
       "      <td>13.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      13\n",
       "n_devices                     180.630137\n",
       "n_users_same_dev               23.013699\n",
       "n_trades                     2210.671233\n",
       "n_put                        1112.972603\n",
       "n_call                       1097.698630\n",
       "n_turbo                       647.479452\n",
       "n_digital                    1563.191781\n",
       "n_Unknown platform            555.301370\n",
       "n_WebGL                        55.986301\n",
       "n_WinGL                      1157.178082\n",
       "n_Desktop Mobile                0.178082\n",
       "n_Android                      32.890411\n",
       "n_IQ Option X                 168.054795\n",
       "n_WEB_MOBILE                   89.958904\n",
       "n_Ios                         150.000000\n",
       "n_Android pwa for iqoption      0.164384\n",
       "n_MacGL                         0.246575\n",
       "n_Ios CFD                       0.328767\n",
       "n_Android Facebook              0.000000\n",
       "n_IQ Option Wallet Web          0.205479\n",
       "n_Android Vertical              0.000000\n",
       "n_CapitalBear.Android           0.000000\n",
       "n_IOS pwa for iqoption          0.178082\n",
       "n_CapitalBear.WebGL             0.000000\n",
       "n_CapitalBear.WinGL             0.000000\n",
       "n_Exnova.WebGL                  0.000000\n",
       "n_CapitalBear.AndroidX          0.000000\n",
       "pnl_total                    9344.598914\n",
       "n_wins                         42.424658\n",
       "total_investment            58429.725992\n",
       "avg_act_level                   1.143996\n",
       "reg_year                     2020.260274\n",
       "n_quest_pnl                   281.452055\n",
       "n_night_quest_pnl              10.000000\n",
       "cluster                        13.000000"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.query('cluster == 13').T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "66c59b7c-f36d-40ce-b140-fe2ca554ac19",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}