Skip to content

clementks/retail_analytic_RFM_Segmentation

Repository files navigation

RFM_Segmentation & Association Mining for Cross-Selling

Predictive DecisionTree Classifier based on Customer RFM Segment_Presentation

Dataset: Typical sales transactions from retail e-commerce store used to determine the customer purchasing behaviour before making RFM (Recency, Frequency & Monetary) analysis

Objective of this presentation using DecisionTree Classifier:

⏉ Recency, Frequency, and Monetary Value (RFM) helps inform customer segmentation in clusters and identifies probability to churn for each cluster.

⏉ Then could devise a targeted marketing campaign and incentivize these customers to extend their contract and continue to procure products or services to ensure customer retention instead of high cost of acquiring new customer.

⏉ Based on customer’s purchase history enables the establishment of churn thresholds for each customer group and assists in constructing a model to predict future churners.

⏉ Make proactive decision to improve customer retention after understanding the propensity of customers to churn.

Model Evaluation Techniques:

 To measure model optimization we used the f1 score. It provides a better measure of incorrectly 
classified cases while accuracy only measures cases that are correctly identified. In addition, the f1 
score is a better metric for an imbalanced class like the one observed in our dataset

Cross-Selling using Association Mining Algorithm

Cross selling is the ability to sell more products to a customer by analyzing the customer’s purchasing trends as well as patterns, which is ubiquitous in both the online and offline worlds. The simplicity and the effectiveness of the idea make it an essential and powerful marketing tool for all types of retailers. The idea of cross selling can be extended to any organization, irrespective of whether it is an online or offline retailer or whether it is selling its products to the end users of whole sellers. In the file that i have attached includes my exploration in association rule-mining using Python Scripting language since it is a useful market basket analysis technique that can be used to determine the potential opportunities for cross selling.

image

image

image

PS: Refer to the "Market Basket Analysis_Customer Segmentation and Association-Mining using Apriori and FP Growth Algorithm.pdf" file in this directory for more details of my data exploration

Citation (Reference):

  1. Richard Farrow, William Trevino, Vitaly Briker, and Brent Allen, “Identifying Customer Churn-in After-market Operations using Machine Learning Algorithms”, Vol. 2 Issue 3

About

Customer RFM Segment_Presentation

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published