Stochastic Text Sentiment Analysis of Modern Chinese Literature Using Fast Fourier Transform
Fengling Shou · 2019
With the rapid expansion of the commentary information, it is difficult to deal with such massive information by manual methods, and it is necessary to rely on computers to assist in acquisition and processing. In this paper, an algorithm based on two-way sentiment analysis is proposed, and an actual system is constructed to perform real-time analysis of the emotional needs of users and recommend literary works. On the one hand, a feature classifier is constructed based on the modern literary audio in the frequency domain. The feature classifier is used to complete the emotional classification of literary works. On the other hand, the real-time text information of users in the social network is obtained to analysis the users based on the fast Fourier transform. Current emotional needs ultimately lead to a list of recommended literary works for users, which allows real-time literary work recommendations based on contextual awareness. The experiments show that the use of personalized recommendation algorithm has higher accuracy, and the user community can obtain a more satisfactory experience.