A collaborative filtering approach using machine learning and business intelligence: a critical review

Muhamad Ibrahim, Najhan and Abiduzzaman, S M and Abdul Raziff, Abdul Rafiez and Shah, Asadullah (2025) A collaborative filtering approach using machine learning and business intelligence: a critical review. International Journal on Perceptive and Cognitive Computing (IJPCC), 11 (1). pp. 41-49. E-ISSN 2462-229X

Preview

PDF - Published Version Download (3MB) | Preview

Abstract

In today's digital context, internet buying has become a common way of consumer behaviour, necessitating the creation of highly personalised recommendation systems. This study provides a critical analysis of a collaborative filtering technique that uses machine learning and business intelligence (BI) to improve e-commerce recommendation systems. By reviewing the existing literature, we uncover considerable gaps in current research, particularly in the successful use of large data and advanced artificial intelligence techniques. Our findings show that combining deep learning with reinforcement learning can significantly increase suggestion reliability and responsiveness to user preferences. Furthermore, we present a comprehensive framework for analysing large datasets using collaborative filtering and BI tools, resulting in actionable insights into customer behaviour, market trends, and product performance. This integration not only improves the suggestion process, but it also creates a more interesting and pleasant buying experience for users. Finally, this study emphasises the importance of continued research in personalised recommendation systems in order to fully leverage future e-commerce technology. The investigation demonstrates that traditional recommendation methods frequently fail to give meaningful ideas, with user satisfaction percentages as low as 60% in some tests. In contrast, our suggested architecture, which integrates collaborative filtering and BI technologies, shows a considerable increase in suggestion accuracy. Specifically, we discovered that combining deep learning techniques with reinforcement learning algorithms enhanced recommendation reliability by 35% while improving user engagement measures by 25%. Furthermore, the incorporation of BI tools improved data visualisation and predictive analytics, allowing e-commerce companies to better understand customer behaviour and market trends. This study emphasises the importance of continued research and innovation in personalised recommendation systems, advocating for a comprehensive approach that leverages the potential of emerging technologies to satisfy consumers' growing expectations in the competitive e-commerce landscape

Item Type:	Article (Journal)
Uncontrolled Keywords:	A Collaborative Filtering, Machine Learning, Business Intelligence
Subjects:	T Technology > T Technology (General) > T10.5 Communication of technical information
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Information and Communication Technology > Department of Information System Kulliyyah of Information and Communication Technology > Department of Information System
Depositing User:	Dr. Najhan Muhamad Ibrahim
Date Deposited:	21 May 2025 14:37
Last Modified:	21 May 2025 14:37
URI:	http://irep.iium.edu.my/id/eprint/121144

Actions (login required)

View Item