• Labeled the news and comments data collected from Yahoo.com and found that 70.8% of the comments’ topics shift.
• Defined five features for comment and then got the best feature combination with classification accuracy of 82.4% using information entropy and SVM classifier.
Sun, Yingcheng, and Kenneth Loparo. "Topic Shift Detection in Online Discussions using Structural Context." In 2019 IEEE 43rd Annual Computer Software and Applications Conference (COMPSAC), vol. 1, pp. 948-949. IEEE, 2019.
Sun, Yingcheng, and Kenneth Loparo