Tensor-based Collaborative Filtering With Smooth Ratings Scale

Abstract

Conventional collaborative filtering techniques don't take into consideration the effect of discrepancy in users' rating perception. Some users may rarely give 5 stars to items while others almost always assign 5 stars to the chosen item. Even if they had experience with the same items this systematic discrepancy in their evaluation style will lead to the systematic errors in the ability of recommender system to effectively extract right patterns from data. To mitigate this problem we introduce the ratings' similarity matrix which represents the dependency between different values of ratings on the population level. Hence, if on average the correlations between ratings exist, it is possible to improve the quality of proposed recommendations by off-setting the effect of either shifted down or shifted up users' rates.

Enhanced Graph Learning for Recommendation via Causal Inference

“…This model combines session information and temporal awareness to overcome the complexity drawbacks of hierarchical models. Marin [3] corrects rating bias caused by evaluation habits by introducing a similarity matrix of evaluations to identify and quantify the variability of different users' evaluation styles. The NCF model proposed by He [4] introduced deep learning techniques into recommendation models for the first time and established a two-tower structure of user embedding and item embedding interaction.…”