Indian Journal of Science and Technology

Abstract

Data mining is the process of arranging and classifying massive data and revealing the related information. Nowadays, data mining is utilized as a highly significant means for managers to understand more precisely the situation of their organization and help with decision making. Utilization of recommender systems is a modern technique to examine the data available in an organization with the use of powerful tools. Challenges involved with the dataset of these systems have intensified the complexities of decision making, such that most effective variables are unknown and the relationship between them is nonlinear and complicated. In such a condition, traditional devices cannot be employed to analyze data and derive knowledge from them; therefore, utilizing matrix analysis and then tensor decomposition has been preferred in recent years compared to other methods. In the present study, time is considered as the third dimension of tensor and an independent factor. The results indicate that using time is effective in reducing estimation error.
Keywords: Data Mining, Recommender Systems, Tensor Decomposition