Rating prediction by exploring user’s preference and sentiment

With the development of e-commerce, shopping on-line is becoming more and more popular. The explosion of reviews have led to a serious problem, information overloading. How to mine user interest from these reviews and understand users’ preference is crucial for us. Traditional recommender systems ma...

Full description

Saved in:
Bibliographic Details
Published in:Multimedia tools and applications 2018-03, Vol.77 (6), p.6425-6444
Main Authors: Ma, Xiang, Lei, Xiaojiang, Zhao, Guoshuai, Qian, Xueming
Format: Article
Language:English
Subjects:
Computer Communication Networks
Computer Science
Data Structures and Information Theory
Multimedia Information Systems
Overloading
Ratings & rankings
Recommender systems
Sentiment analysis
Social factors
Social networks
Special Purpose and Application-Based Systems
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With the development of e-commerce, shopping on-line is becoming more and more popular. The explosion of reviews have led to a serious problem, information overloading. How to mine user interest from these reviews and understand users’ preference is crucial for us. Traditional recommender systems mainly use structured data to mine user interest preference, such as product category, user’s tag, and the other social factors. In this paper, we firstly use LDA+Word2vec model to mine user interest. Then, we propose a social user sentimental measurement approach. At last, three factors, including user topic, user sentiment and interpersonal influence, are fused into a recommender system (RS) based on probabilistic matrix factorization. We conduct a series of experiments on Yelp dataset, and experimental results show the proposed approach outperforms the existing approaches.
ISSN:1380-7501
1573-7721
DOI:10.1007/s11042-017-4550-z