A deep recommendation model of cross-grained sentiments of user reviews and ratings

作者：

Highlights：

• Proposes a deep learning recommendation model which integrates textual review sentiments and rating matrix.

• The proposed model combines cross-grained sentiment of reviews and user-item rating-based matrix factorization.

• The scheme extracts both fine-grained sentiments at the subsentence level and coarse-grained sentiments at the sentence level of reviews.

• Experimental results show that the proposed model achieved better prediction results than other existing models.

摘要

•Proposes a deep learning recommendation model which integrates textual review sentiments and rating matrix.•The proposed model combines cross-grained sentiment of reviews and user-item rating-based matrix factorization.•The scheme extracts both fine-grained sentiments at the subsentence level and coarse-grained sentiments at the sentence level of reviews.•Experimental results show that the proposed model achieved better prediction results than other existing models.

论文关键词：Review text,Rating matrix,Cross-grained sentiment analysis,Recommendation model

论文评审过程：Received 21 July 2021, Revised 24 November 2021, Accepted 1 December 2021, Available online 22 December 2021, Version of Record 22 December 2021.

论文官网地址：https://doi.org/10.1016/j.ipm.2021.102842