A New Similarity Method to Optimize Business in the Online Stores Using the Rating Time Technology

These days, Emergence of e-commerce web sites is one of the important consequences of the Internet in modern times, but products data is growing exponentially. In such environment, customers face a problem in finding optimized information among huge data bases about the items or desired products. In order to assist buyers, large e-commerce companies are planning to introduce their own recommender systems to help their customers in making a better choice among the items. Due to high percentage error , a basic method to build such systems is not usually being applied. In this essay, two methods have been suggested in order to improve recommendations in recommender systems. Collaborative filtering method is one of the most successful methods used in the system, but using this method that it has common problem the increasing number of users and products, therefore system do not inability to request the requirement of cold start and data sparsity. Two methods have been suggested in order to improve recommendations in recommender systems. To resolve this problem, a new method has been introduced in which by integrating  rating time by Pearson also combining semantic technology with social networks offers a solution to reduce issues such as "cold start" and generally "data sparsity" in recommender systems. The result of simulating showed that the proposed approach provided better performance and more accurate predictions in addition of more consistent with user preferences.

A New Similarity Method to Optimize Business in the Online Stores Using the Rating Time Technology Keywords:

Data sparsity , Internet store , Recommender systems , Users rating time