Title: An efficient collaborative recommender system for textbooks using silhouette index and K-means clustering technique
Authors: Dinesh Kumar Yadav; Rati Shukla; Vikash Yadav
Addresses: Department of Computer Science and Engineering, Delhi Technological University, Delhi, India ' Department of GIS Cell, Motilal Nehru National Institute of Technology, Allahabad, India ' Department of Computer Science and Engineering, ABES Engineering College, Ghaziabad, India
Abstract: A recommender system provides a great platform for filtering of information and various knowledge-based management systems. They provide very good recommendation to the users so that they are able to predict the quality of the product in e-commerce. In today's research, it is very difficult to predict the accurate information regarding online products. In this research, we are going to introduce textbook-based recommender systems which uses the silhouette index and k-means clustering technique to predict the ratings of the textbooks available online based on its previous data. Initial positions of clusters are obtained and classifying the clusters by similarity of users are done by k-means clustering techniques. Our proposed recommender systems are able to predict much improved results rather than the other available state-of-the-art methods. Efficiency and performance of this newly developed recommender systems is enhanced when compared with existing systems. All experiments are done using publicly available BX-Book-Ratings datasets and achieved the MAE of 0.63 which is best among other state-of-the-art recommendation systems.
Keywords: recommender system; silhouette index; K-means clustering; biometric; machine learning; root mean squared error; mean absolute error.
International Journal of Advanced Intelligence Paradigms, 2021 Vol.19 No.2, pp.233 - 242
Received: 29 Nov 2018
Accepted: 10 Dec 2018
Published online: 14 May 2021 *