Open AccessEmotional analysis System of book review based on Neural network
Author(s) -
Yanxiong Sun,
Yeli Li,
Qingtao Zeng,
Yuning Bian
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/563/4/042012
Subject(s) - publishing , credibility , publication , computer science , sentiment analysis , feeling , world wide web , artificial intelligence , advertising , psychology , political science , business , law , social psychology
Contemporary society, is an era of the Information explosion. The explosive growth of book publishing makes readers choose a good book, and publishers choose to publish a book with a market, which has become a very important topic. Aiming at the non-effective analysis of the current book publishers to obtain the real feelings of the users on the book, this paper puts forward that the neural network as the main algorithm is enough to classify the user’s comments emotionally. In the collection of data, through the award-winning books and the two-star score under the book review, take different weights of the calculation, the emotional words of the eigenvectors to plump. And after the book evaluation is pre-processed, the user’s credibility parameters are added to the input word vector. As a result, the accuracy of classification and judgment of the whole system is improved, the consumption of human, material and financial resources is reduced, and the resource can be invested in how to improve the level of publications, which will add to the cause of book publishing in China.