Faerie Entity Extraction

This is an implementation of faeire entity extraction, a dictionary-baesd entity extraction.

Installation and dependencies

pip install faerie

Usage

Run it in console

faerie sampledictionary.json sampledocuments.json sampleconfig.json

Or you can import and use it in your python code

import faerie
faerie.run("sampledictionary.json","sampledocuments.json","sampleconfig.json")

Example input and dictionary file format

{
  "Country": "united states",
  "State": "georgia",
  "AdministrativeArea": "douglas county",
  "name": "Woods Valley",
  "id": "4231869"
}

Example config file format

{
  "dictionary": {
    "id_attribute": "id",
    "value_attribute": [
      "name"
    ]
  },
  "document": {
    "id_attribute": "id",
    "value_attribute": [
      "name"
    ]
  },
  "token_size": 2,
  "threshold": 0.8
}

Arguments:

Required

• documents a json file for extracting entities

And

• dictionary a json file providing entities to extract

Optional

• config a json file contains fields, tokensize and threshold

Output

{
  "entities": {
    "4231869": {
      "value": "Woods Valley",
      "candwins": [
        {
          "start": 64,
          "score": 11,
          "end": 74
        },
        {
          "start": 65,
          "score": 10,
          "end": 74
        },
        {
          "start": 196,
          "score": 10,
          "end": 206
        }
      ]
    }
  },
  "document": {
    "id": "4231868",
    "value": "The choice of San Diego golf enthusiasts and beginners alike is Woods Valley Golf Club. Stunning tree-lined fairways and beautiful rolling slopes highlight the 18-hole championship golf course at Woods Valley. Designed with an emphasis on strategy and shot-making, the course winds through the community, following the natural contours of the countryside."
  }
}