Development of an Intelligent E-Commerce System Using Product Recommendation Technology

Abstract

This research aimed to (1) develop an intelligent E-Commerce system using product recommendation technology, (2) evaluate the performance of the developed algorithm, and (3) examine the impact of the system on consumers’ purchasing behavior and technology acceptance. The sample consisted of 250 participants selected through purposive sampling from users who had experience in online shopping. The research instruments included a prototype system developed using a hybrid recommendation approach, combining collaborative filtering and content-based filtering techniques, and a five-point Likert scale questionnaire. Data were analyzed using descriptive statistics, t-test, and multiple regression analysis.

           The results revealed that (1) the developed intelligent E-Commerce system using a hybrid recommendation approach, which integrates collaborative filtering and content-based filtering techniques, performed effectively. The system achieved a precision value of 0.82, a recall value of 0.79, and an F1-score of 0.80, indicating that the system was able to recommend products with a high level of accuracy. (2) The evaluation results from 250 users indicated that perceived usefulness, perceived ease of use, and overall satisfaction with the system were at a high level. Moreover, the use of the product recommendation system significantly increased purchase decisions and repurchase rates at the .05 level of statistical significance.