This paper leverages insights from my previous works to analyze and predict customer behavior in different areas using data mining and machine learning techniques. The research focuses on identifying and interpreting customer preferences, purchasing patterns and risk factors for customer churn to develop predictive models that support decision making. By analyzing various datasets such as tabular data, clickstream data, and digital interactions, this study aims to provide a comprehensive framework for extracting actionable insights and enhancing the accuracy and transparency of customer behavior predictions. The proposed approach enables businesses to anticipate customer behavior, personalize marketing strategies, and improve the explainability and effectiveness of customer behavior prediction models. Future work will incorporate advanced visualization techniques and various data structures to further improve the model’s transparency and predictive capability.

In the rapidly evolving e-commerce industry, the ability to select high-quality data for model training is essential. This study introduces the High-Utility Sequential Pattern Mining using SHAP values (HUSPM-SHAP) model, a utility mining based active learning strategy, to tackle this challenge. We found that the parameter settings for positive and negative SHAP values affect the mining results of the model, introducing a key consideration into the active learning framework. Unlike traditional SHAP, which evaluates individual elements, HUSPM-SHAP utilizes SHAP values in combination with HUSPM to identify the valuable of high-utility sequential patterns for improving prediction models. In experiments to predict behaviors that actually lead to purchases, the developed HUSPM-SHAP model shows its superiority in different scenarios. The model’s ability to reduce labeling requirements while maintaining high predictive performance is highlighted. Our results show that the model is able to refine the processing of e-commerce data and lead to optimized, cost-efficient prediction modeling.

Customer churn is a most crucial challenge faced by various industries, especially for the telecommunications sector since it causes over 5 - 6 times cost to keep customers. Thus, customer churn prediction has experienced substantial expansion. While a lot of prediction models are becoming increasingly accurate, the capability to interpret these models and perform causal analysis has become a new challenge. Identifying the key reasons or patterns that truly influence customer churn is vital for enabling industries to make significant improvements. This study then introduced a comprehensive framework. That integrates causal graphs and explainable models such as SHAP and Shapley flow, and concepts from system dynamics to explore customer churn patterns and leverage points. Experimental results indicates that the designed model is more explainable and interpretable to explore the customer behaviors for customer churn prediction.