Survey on Users Ranking Pattern based Trust Model Regularization in Product Recommendation

Sneha U,Liji Samuel

doi:10.31142/ijtsrd11302

Abstract

Survey on Users Ranking Pattern based Trust Model Regularization in Product Recommendation

Highlights

The recommendation system helps the user to potentially reduce the reliance on search algorithms by improving the discoverability of the items because they connect related information to users who may be hard to find
The recommendation system is mainly used in many fields such as e-commerce, marketing, financial services, and personalized holidays, etc
The recommended technologies are roughly divided into content-based filtering technologies, collaborative filtering technologies, and hybrid recommendation systems

Summary

INTRODUCTION

The recommendation system helps the user to potentially reduce the reliance on search algorithms by improving the discoverability of the items because they connect related information to users who may be hard to find. Most e-commerce sites such as Myntra.com and social networking sites such as facebook.com have such a recommendation system These systems serve two important tasks (1), helping users to process excess information by providing users with appropriate advice (2), and helping users to earn more profits by selling more products. Content-based filtering strategy: Use user preference profiles and project description information about historical transactions. Neighbor Technology: The recommendation system can use trusted neighbors because it can improve the accuracy of the recommendation system, coverage, and system performance This method works well when the sparsely distributed rating data is sparse, but when the ratings are diversified, most of the strategies used in this paper do not apply. It provides an effective global optimization scheme, but it has not been tested with large data sets that are related to implicit feedback It integrates implicit user feedback into the neighborhood model and latent factor model (svd)

LITERATURE SURVEY

CONCLUSION