Abstract
As user-generated content increasingly proliferates through social networking sites, our lives are bombarded with ever more information, which has in turn has inspired the rapid evolution of new technologies and tools to process these vast amounts of data. Semantic and sentiment analysis of these social multimedia have become key research topics in many areas in society, e.g., in shopping malls to help policymakers predict market trends and discover potential customers. In this light, this study proposes a novel method to analyze the emotional aspects of Chinese vocabulary and then to assess the mass comments of the movie reviews. The experiment results show that our method 1. can improve the machine learning model by providing more refined emotional information to enhance the effectiveness of movie recommendation systems, and 2. performs significantly better than the other commonly used methods of emotional analysis.
Highlights
Thanks to the booming development of social media in recent years, we can acquire vast amounts of user-generated data
To summarize, according to the above experiment results, the Refined Distributed Emotion Vector (RDEV) can improve the performance of Support Vector Machines (SVM), TextCNN, and long short-term memory (LSTM) by providing more detailed emotional information to enhance the effectiveness of sentiment classification, and performs significantly better than the other commonly used methods of emotional analysis
This paper proposes a novel approach to sentiment analysis research, which aims at predicting trends in public opinions using sophisticated methods of emotion analysis
Summary
Thanks to the booming development of social media in recent years, we can acquire vast amounts of user-generated data. Research on opinion and sentiment analysis has been widely adopted by businesses They collect social media information from multiple sources as a means to determine the trends of public opinion so that they respond quickly and create competitive advantage [1]. Cross-domain sentiment classification algorithms are highly desirable to reduce domain dependency and manually labeling costs In light of these considerations, this work proposes a fine-grained sentiment analysis method based on Valence and Arousal prediction. We further utilize this method to imbue words with sentiment meanings and present a novel document representation model.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
