Customer-dna 0.1.2

A customer analysis package for data

Customer Segmentation, CLTV, and MBA Analysis

This package performs Customer Segmentation, Market Basket Analysis (MBA), and Customer Lifetime Value (CLTV) calculation with features for predicting Customer Attrition and generating Next Best Product Recommendations.

Features

• Customer Segmentation: Clusters customers based on purchasing behavior using clustering models.
• Market Basket Analysis (MBA): Performs market basket analysis to understand product association patterns.
• CLTV Calculation: Estimates customer lifetime value using BG/NBD and Gamma-Gamma models.
• Customer Attrition Prediction: Identifies customers at risk of leaving the company.
• Next Best Product Recommendation: Recommends products based on customer purchase behavior.

Installation

1. Clone the repository:

   pip install Customer_dna
   

2. Install the required dependencies:

   pip install -r requirements.txt
   

Usage:

This package provides a main() function that runs customer segmentation, market basket analysis, CLTV calculation, attrition prediction, and product recommendation.

import pandas as pd
from customer_dna.main import main

# Load your dataset
df = pd.read_csv("your_dataset.csv")

# Call the main function
recommendations = main(df, sample_size=10000, top_n=5)

# Output the recommendations
print(recommendations)

Data Requirements

Your input data should have the following columns (or their synonyms as detected by the package):

• InvoiceNo: Invoice number of the transaction.
• InvoiceDate: Date of the transaction.
• Description: Product description.
• Quantity: Quantity purchased.
• UnitPrice: Price of the product.
• CustomerID: Unique customer identifier.
• Country: Country of the customer.

Key Functions

1. Customer Segmentation (customer_segmentation()): Segments customers based on purchasing behavior using PyCaret’s clustering models.

2. Market Basket Analysis (MBA) (mba()): Analyzes product associations in different customer segments using mlxtend.

3. CLTV Calculation (process_and_visualize_clv()): Computes customer lifetime value using the BG/NBD and Gamma-Gamma models.

4. Customer Attrition Prediction (predict_customer_attrition()): Predicts customers likely to churn based on the CLTV output.

5. Product Recommendation (next_best_product_recommendation()): Recommends the next best products for each customer based on previous purchasing patterns.

Required Modules

pandas==1.3.3
numpy==1.21.2
matplotlib==3.4.3
seaborn==0.11.2
scikit-learn==0.24.2
scipy==1.7.1
pycaret==2.3.3
lifetimes==0.11.3
mlxtend==0.19.0
plotly==5.3.1
networkx==2.6.3