Arousal Prediction of News Articles in Social Media

Kumar, Nagendra and Singh, Manish and Suryamukhi, K et. al. (2017) Arousal Prediction of News Articles in Social Media. In: International Conference on Mining Intelligence and Knowledge Exploration, 13-15 December 2017, Hyderabad, India.

Full text not available from this repository. (Request a copy)


At present, news channels are using social media to disseminate news to a large audience. These news channels try to convey the news in such a way that attracts more user interaction in the form of views, likes, comments, and shares. In online news, one of the important factors in getting higher user interaction is to recommend the top news articles that would attract more number of users to give opinion, especially in the form of comments. When a news article starts getting many comments, it automatically attracts other readers to participate in the discussion. We say that a news article has “high-arousal” content if it can attract more number of comments from users. When a new news article is written it has no user interaction information, such as number of views, likes, shares or comments. In this paper, our aim is to predict news articles which have higher potential to generate high-arousal. In other words, they would attract a large number of users to give opinion in the form of comments. Unlike previous studies, we predict the arousal of news articles prior to their release, which brings the possibility of appropriate decision making to modify the article content or its ranking in audience newsfeed. We generate multiple features from the content of news articles and show that our best set of features can predict the arousal with an accuracy of 81%. We perform our experiments on social media page of CNN news channel, containing four years of data with 33,324 news articles and 226.83 million reactions.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Singh, Manish
Item Type: Conference or Workshop Item (Paper)
Subjects: Computer science
Divisions: Department of Computer Science & Engineering
Depositing User: Team Library
Date Deposited: 17 May 2019 05:53
Last Modified: 17 May 2019 05:53
Publisher URL:
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 5218 Statistics for this ePrint Item