Automatic Detection of Clickbait Headlines Using Semantic Analysis and Machine Learning Techniques

Abstract

Clickbait headlines are misleading headiness designed to attract attention and entice users to click on the link. Links can host malware, trojans and phishing attacks. Clickbaiting is one of the more subtle methods used by hackers and scammers. For these reasons, clickbait is a serious issue that must be addressed. This paper presents a method for identifying clickbait headlines using semantic analysis and machine learning techniques. The method involves analyzing thirty unique semantic features and exploring six different machine learning classification algorithms individually and in ensemble forms. Results show that the top models have an accuracy of 98% in classifying clickbait headlines. The proposed models can serve as a template for developing practical applications to detect clickbait headlines automatically.