Leveraging machine learning for churn prediction: techniques, challenges and integration with customer lifetime value

Carucci, Lorenzo (A.A. 2023/2024) Leveraging machine learning for churn prediction: techniques, challenges and integration with customer lifetime value. Tesi di Laurea in Data visualization, Luiss Guido Carli, relatore Blerina Sinaimeri, pp. 85. [Master's Degree Thesis]

[img]
Preview
PDF (Full text)
Download (2MB) | Preview

Abstract/Index

Literature review. The costumer churn problem. Key terms and definitions. Introduction to churn classification models. Introduction to gradient boosting techniques. LightGBM: an optimized and scalable gradient boosting framework. Unbalanced class problem. SMOTE approach. Data overview and methods. Approach explanation. Input data. Data transformation. Data migration and cloud overview. CAR dataset. Preprocessing and model overview. Data processing steps for churn analysis. Data preprocessing: correlation matrix considerations. Split and SMOTE operations. Algorithms explanation. Result. Tuning. Tuning result. Model implementation. Model inference: application for predictions on new data. Churn probability and targeted retention strategies based on risk levels.

References

Bibliografia. pp. 81-84.

Thesis Type: Master's Degree Thesis
Institution: Luiss Guido Carli
Degree Program: Master's Degree Programs > Master's Degree Program in Data Science e Management (LM-91)
Chair: Data visualization
Thesis Supervisor: Sinaimeri, Blerina
Thesis Co-Supervisor: Martino, Alessio
Academic Year: 2023/2024
Session: Autumn
Deposited by: Alessandro Perfetti
Date Deposited: 15 Apr 2025 11:00
Last Modified: 15 Apr 2025 11:00
URI: https://tesi.luiss.it/id/eprint/41812

Downloads

Downloads per month over past year

Loading...

Repository Staff Only

View Item View Item