Volume 12, Issue 1 (3-2020)                   itrc 2020, 12(1): 42-55 | Back to browse issues page

XML Print


Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

Mohamadrezaei R, Ravanmehr R. Trend Detection and Prediction in Blogosphere based on Sentiment Analysis using PSO and Q-Learning. itrc 2020; 12 (1) :42-55
URL: http://journal.itrc.ac.ir/article-1-453-en.html
1- Department of Computer Engineering, Central Tehran Branch, Islamic Azad University Tehran, Iran
2- Department of Computer Engineering, Central Tehran Branch, Islamic Azad University Tehran, Iran , r.ravanmehr@iauctb.ac.ir
Abstract:   (2204 Views)
The blogosphere is an effective communication platform where users publish and exchange their opinions. By analyzing user behavior, current and future trends of a community can be discovered. The proposed model for processing the social data of users first extracts related sentiments of weblog comments. An improved PSO algorithm is then employed to detect the trend of users in the TRDT (TRend DeTection) phase. By the discovery of trends at a reasonable time and appropriate precision, this model predicts future trends of the blogosphere using the Q-learning algorithm in the TRPT (TRend PredicTion) phase. Given the ever-increasing processing requirements and a huge volume of data, our approach provides a distributed processing/storage platform for TRDT and TRPT phases. The precision and performance of the proposed model in the TRDT phase are measured by the Chi-squared standard test. Moreover, the evaluation of the TRPT phase shows the comparable precision of the proposed approach with real-world scenarios such as the Netflix predictive system. 
Full-Text [PDF 1239 kb]   (717 Downloads)    
Type of Study: Research | Subject: Information Technology

Add your comments about this article : Your username or Email:
CAPTCHA

Send email to the article author


Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.