Time Based Collaborative Recommendation System by using Data Mining Techniques

Abstract

Recommendation of appropriate product to the specific user is becoming the key to ensuring the continued success of E-commerce. Today, many E-commerce systems adopt various recommendation techniques, e.g., Collaborative Filtering (abbreviated as CF)-based technique and Structural Balance Theory-based Recommendation (i.e., SBT-Rec) technique to realize product item recommendation. Overall, the present CF recommendation and as per suggested SBT can perform very well, if the target user owns similar friends (user-based CF) and Structural Balance Theory-based Recommendation (i.e., SBT-Rec) for we first look for the target user’s dissimilar “enemy” (i.e., antonym of “friend”), and furthermore, we look for the “possible friends” of E-commerce target user, according to “enemy’s enemy is a friend” rule of Structural Balance Theory or the product items purchased and preferred by target user own one or more similar product items (item-based CF). Here both the systems depends on friends and enemies if we are not getting friends or enemies then. So to improve Recommender system we propose a time-aware profile based collaborative Recommendation algorithm. In this algorithm, we will consider only recently submitted ratings and positive reviews to evaluate products quality. Along with this, we propose a novel recommender system in which user will give his requirement about any product as input, and depending on that input we will recommend most appropriate products according to the customer’s requirement and ratings given by other customers. Only recent ratings will be considered by the system. Our proposed system will meet personalized product item recommendation requirements in E-commerce and time-aware rating consideration to evaluate current product quality.