CASE STUDY –E-Commerce Platform Losing Customer from Alarming Rates
Devanshi Dembda Student
Dr Divya Lakhani Associate Professor
Sadhu Vaswani Institute of Management Studies for Girls, Affiliated to Savitribai Phule Pune University
6, Koregaon Park, Pune 411001 [Maharashtra, India]
1lakhani.divya@svims-pune.edu.in
2dembda.devanshi@svims-pune.edu.in
Some customer segments are more prone to churn and seek to identify these groups Proactively.working as a data scientist and while using the given data said I have performed following task scientist tasked with developing an AI-powered solution to Predict customer churn and propose strategies to retain valuable customers.Data: The dataset is available to download at
https://www.kaggle.com/datasets/ankitverma2010/ecommerce-customer-churn-analysisandprediction
Identify and justify the specific AI approach best suited for this problem.
To address the challenge of predicting customer churn in an E-Commerce platform and proposing strategies to retain valuable customers, the most suitable AI approach would be Machine Learning, specifically Supervised Learning algorithms like Logistic Regression, Random Forest, or Gradient Boosting.
Problem Identification:
The problem involves predicting customer churn, which is a classic classification problem where the goal is to predict whether a customer will churn or not based on historical data and customer attributes.
AI Approach
Justification:
• Supervised Learning: Since the dataset is labeled with churn information, supervised learning models can be trained to predict future churn.
• Logistic Regression: Suitable for binary classification tasks like churn prediction, providing probabilities and feature importance.
• Random Forest: Effective for handling large datasets with many features and capturing non-linear relationships.
• Gradient Boosting: Known for its high predictive accuracy and ability to handle complex interactions in the data. By utilizing these AI approaches, the E-Commerce platform can build predictive models to identify customer segments prone to churn, enabling them to implement targeted retention strategies and reduce customer attrition effectively.
a program to perform the key steps involved in developing your chosen AI Model for customer churn prediction.
To develop an AI model for customer churn prediction, we will follow the key steps involving data preprocessing, training, and testing data split, and fitting a model. Below is a Python program that demonstrates these steps using the Telco Customer Churn dataset available on Kaggle.
