Open AccessThe Quality Evaluation of Online Book Reviews Based on Graph Neural Network
Author(s) -
Can Zhang,
Wei Liu,
Qi Mu,
Deshan Zhang,
Fancheng Meng
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2010/1/012080
Subject(s) - computer science , the internet , graph , artificial neural network , quality (philosophy) , reading (process) , data science , artificial intelligence , machine learning , information retrieval , data mining , world wide web , theoretical computer science , philosophy , epistemology , political science , law
With the development of the Internet, the way people read has changed, the readers are more and more willing to share their reading opinions via the reading platforms. However, as the continuous increase in the number of users and reviews on the book platforms, the review data is growing exponentially. It is difficult for people to find useful information quickly in the massive reviews. Therefore, using automated methods to identify high-quality reviews among a large number of book reviews has become a hot topic in current research. This paper studies a quality evaluation method of book reviews based on graph convolution neural network, which introduces the LDA document topic generation model to add topic nodes on the basis of review and word nodes, builds the relationship between the three nodes as edges to generate heterogeneous graph. The graph is sent to a two-layer GCN model for training. In the last, this article did experimental evaluation and analysis based on the book review data of non-commercial platforms. The results show that the proposed method is superior to the machine learning method and the traditional neural network model.