.png){kind=icon}
.png){kind=icon}
Return
Recommendation algorithm combining ratings and comments
.png){kind=icon}
OA
AI
DOI：10.1016/j.aej.2021.04.056.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The rating data of recommendation systems are too sparse, and many existing studies introduce review information to alleviate this phenomenon, but the connection between reviews and the importance of each review is not fully considered in the feature extraction of the information of reviews. To address this problem, this paper proposes the Local-Global Awareness Attention Model (LGAA) to model the comment information. Through the local attention mechanism, the neighborhood attention score is calculated by fully considering the association between each comment and other comments; then the global attention mechanism is used to calculate the global attention score of each comment and weight the sum of all comments, and fuse the scoring information to form a feature vector; finally, it is fed into a factorization machine and a fully connected network for scoring prediction. The results show that LGAA can achieve better performance than the benchmark algorithm. (C) 2021 THE AUTHORS. Published by Elsevier BV on behalf of Faculty of Engineering, Alexandria University.
Keywords:
Attention
Transformer
Factorization machine
Matrix factorization
Rating prediction
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6.8
Papers:
6.3K
Citations:
2.6W