DataScience-MachineLearning_Project

This is my Data Science and Machine Learning Portfolio Project "Customer Segmentation and Churn Prediction for a Telecom Company"

Problem Statement

A telecommunications Company wants to better understand its customer base and identify factors contributing to customer churn. The company wants to segment its customers based on their behaviour and demographics and develop a predictive model to identify customers at risk of churn. By doing so, they aim to proactive measures to retain customers and optimize their market strategies.

Goals

1. Data Preprocessing and Exploration:

• Handle missing values, outliers, and perform necessary data cleaning steps.
• Explore the Dataset to understand the distribution of variables and identify potential correlations.

2. Customer Segmentation using K-Means Clustering:

• Apply K-Means clustering algorithm to segment customers based on their usage patters, demographics or other relevant features.
• Visualize the clusters to gain insights into different customer segments.

3. Feature Engineering:

• Create additional features such as average monthly usage, tenure, and total charges.
• Encode categorical variables appropriately such as one-hot encoding for categorical demographic features.

4. Principal Component Analysis (PCA):

• Use PCA to reduce Dimensionality of the dataset and identify the most important features contributing to customer churn.

5. Churn Prediction using Logistic Regression:

• Build a logistic regression model to predict customer churn based on the segmented customer data and derived features.
• Split the dataset into training and testing sets and evaluate the model's performance using suitable metrics like accuracy, precision, recall and F1-Score.

Data:

IBM Dataset: Telco Company https://www.kaggle.com/datasets/yeanzc/telco-customer-churn-ibm-dataset A fictional telco company that provided home phone and Internet services to 7043 customers in California in Q3.

Analysis

6. Model Evaluation and Interpretation:

• Assess the performance of the churn prediction model and interpret the results to gain insights into factors influencing customer churn.
• Identify the most significant features contributing to churn based on the logistic regression and coefficients.

7. Reccomendations and Actoinable Insights:

• Provide Actionable Insights to the telecom company based on the customer segmentation and churn prediction results.
• Suggest targetted retention strategies for different customer segments to minimize churn and maximize customer satisfaction.