Predicting Purchasing Behavior on E-Commerce Platforms: A Regression Model Approach for Understanding User Features that Lead to Purchasing  

作　　者：Abraham Jallah Balyemah Sonkarlay J. Y. Weamie Jiang Bin Karmue Vasco Jarnda Felix Jwakdak Joshua Abraham Jallah Balyemah;Sonkarlay J. Y. Weamie;Jiang Bin;Karmue Vasco Jarnda;Felix Jwakdak Joshua(College of Computer Science and Engineering, Hunan University, Changsha, China;Department of Health Inspection and Quarantine, Xiangya School of Public Health, Central South University, Changsha, China)

机构地区：[1]College of Computer Science and Engineering, Hunan University, Changsha, China [2]Department of Health Inspection and Quarantine, Xiangya School of Public Health, Central South University, Changsha, China

出　　处：《International Journal of Communications, Network and System Sciences》2024年第6期81-103,共23页通讯、网络与系统学国际期刊（英文）

摘　　要：This research introduces a novel approach to improve and optimize the predictive capacity of consumer purchase behaviors on e-commerce platforms. This study presented an introduction to the fundamental concepts of the logistic regression algorithm. In addition, it analyzed user data obtained from an e-commerce platform. The original data were preprocessed, and a consumer purchase prediction model was developed for the e-commerce platform using the logistic regression method. The comparison study used the classic random forest approach, further enhanced by including the K-fold cross-validation method. Evaluation of the accuracy of the model’s classification was conducted using performance indicators that included the accuracy rate, the precision rate, the recall rate, and the F1 score. A visual examination determined the significance of the findings. The findings suggest that employing the logistic regression algorithm to forecast customer purchase behaviors on e-commerce platforms can improve the efficacy of the approach and yield more accurate predictions. This study serves as a valuable resource for improving the precision of forecasting customers’ purchase behaviors on e-commerce platforms. It has significant practical implications for optimizing the operational efficiency of e-commerce platforms.This research introduces a novel approach to improve and optimize the predictive capacity of consumer purchase behaviors on e-commerce platforms. This study presented an introduction to the fundamental concepts of the logistic regression algorithm. In addition, it analyzed user data obtained from an e-commerce platform. The original data were preprocessed, and a consumer purchase prediction model was developed for the e-commerce platform using the logistic regression method. The comparison study used the classic random forest approach, further enhanced by including the K-fold cross-validation method. Evaluation of the accuracy of the model’s classification was conducted using performance indicators that included the accuracy rate, the precision rate, the recall rate, and the F1 score. A visual examination determined the significance of the findings. The findings suggest that employing the logistic regression algorithm to forecast customer purchase behaviors on e-commerce platforms can improve the efficacy of the approach and yield more accurate predictions. This study serves as a valuable resource for improving the precision of forecasting customers’ purchase behaviors on e-commerce platforms. It has significant practical implications for optimizing the operational efficiency of e-commerce platforms.

关 键 词：E-Commerce Platform Purchasing Behavior Prediction Logistic Regression Algorithm 

分 类 号：F72[经济管理—产业经济]