Detecting "Clickbait" News on Social Media Using Machine Learning Algorithms

Genc S., SÜRER E.

27th Signal Processing and Communications Applications Conference (SIU), Sivas, Türkiye, 24 - 26 Nisan 2019 identifier identifier


Clickbait, which has become very common in social media in recent years, is a technique which uses exaggerated and unreal headlines in order to manipulate people and attract them to their websites. Since the content mentioned in the title is not presented in the main text or the content of text is low-quality, clicked on links often disappoint people. In this study, we attempt to detect clickbaits in Turkish news using Twitter posts. For this purpose, headlines of news were collected from Twitter accounts of Limon Haber(1) and Spoiler Haber(2) for clickbait data and from Twitter accounts of Evrensel Newspaper(3) and Diken Newspaper(4) for non-clickbait data. Experimental results on news headlines show that using an artificial neural network, our model performs for clickbait detection with an accuracy of 0.91 with an F1-score of 0.91-which is the highest score in Turkish data sets.