Authors : Imeh Umoren, Onukwugha Gilean and Juliet Odii
Abstract: As the amount of information available to users on the internet increases geometrically, several approaches are required to assist the user in finding and retrieving relevant information. Intelligent agents with the capacity to learn user’s profile towards efficient sentiment analysis are one solution to this problem. Collaborative Filtering (CF) is one of the most successful recommended approaches used in academia and industry for making automatic predictions (filtering) about the interests of a user by collecting preferences or taste information from many users (collaboration). This work applies Memory-Based CF using Jaccard similarity algorithm in electronic commerce to develop a recommendation system for analyzing user data and extracting user information for accurate predictions of user preferences based on user’s behavior in a Business-to-Consumer (B2C) E-commerce store. The results outcome indicates that CF as a classical method of information retrieval can be used in helping people deal with information overload as the technique reduces the time spent searching for relevant information and also increases the accuracy of retrieval. Furthermore, the results from predictions of user’s interests through recommendation lists are useful for enhance customer’s loyalty and higher marketing rates.
Imeh Umoren, Onukwugha Gilean and Juliet Odii, 2021. Implementing Multimedia Information Retrieval using Memory-Based Collaborative Filtering. Asian Journal of Information Technology, 20: 60-76.