[Paper Review] Stop clickbait: detecting and preventing clickbaits in online news media

Abhijnan Chakraborty, Bhargavi Paranjape+2 more|arXiv (Cornell University)|Aug 18, 2016

Misinformation and Its Impacts12 references200 citations

TL;DR

This paper proposes an automated clickbait detection system integrated into a browser extension that identifies sensationalist headlines and warns or blocks them based on user preferences. The system achieves 93% accuracy in detecting clickbaits and 89% accuracy in personalized blocking through offline and online experiments across multiple news sites.

ABSTRACT

Most of the online news media outlets rely heavily on the revenues generated from the clicks made by their readers, and due to the presence of numerous such outlets, they need to compete with each other for reader attention. To attract the readers to click on an article and subsequently visit the media site, the outlets often come up with catchy headlines accompanying the article links, which lure the readers to click on the link. Such headlines are known as Clickbaits. While these baits may trick the readers into clicking, in the long-run, clickbaits usually don't live up to the expectation of the readers, and leave them disappointed. In this work, we attempt to automatically detect clickbaits and then build a browser extension which warns the readers of different media sites about the possibility of being baited by such headlines. The extension also offers each reader an option to block clickbaits she doesn't want to see. Then, using such reader choices, the extension automatically blocks similar clickbaits during her future visits. We run extensive offline and online experiments across multiple media sites and find that the proposed clickbait detection and the personalized blocking approaches perform very well achieving 93% accuracy in detecting and 89% accuracy in blocking clickbaits.

Motivation & Objective

• To automatically detect clickbait headlines in online news media that mislead readers with exaggerated or sensationalist language.
• To develop a browser extension that warns users when they encounter potential clickbaits.
• To enable personalized blocking of unwanted clickbait content based on user preferences.
• To evaluate the effectiveness of detection and blocking mechanisms across diverse news websites.
• To improve user experience by reducing exposure to misleading headlines that fail to deliver on content promises.

Proposed method

• The system uses natural language processing techniques to analyze headline text and detect linguistic cues associated with clickbait, such as emotional language, curiosity gaps, and exaggerated claims.
• A machine learning model is trained on a dataset of labeled headlines to classify whether a headline is clickbait or not.
• The browser extension integrates the detection model to analyze headlines in real time as users browse news sites.
• Users can flag or block specific clickbaits, and the system learns from these choices to automatically block similar headlines in future visits.
• The system employs a feedback loop to personalize blocking behavior based on individual user preferences.
• Offline and online experiments are conducted across multiple media sites to validate detection and blocking performance.

Experimental results

Research questions

• RQ1 How accurately can automated systems detect clickbait headlines in online news media?
• RQ2 Can a browser extension effectively warn users about potentially misleading clickbait headlines in real time?
• RQ3 To what extent can personalized blocking based on user feedback reduce exposure to unwanted clickbaits?
• RQ4 How do detection and blocking performance vary across different news websites and content types?
• RQ5 What is the impact of user-driven feedback on improving the accuracy of clickbait blocking over time?

Key findings

• The clickbait detection system achieves 93% accuracy in identifying sensationalist headlines across multiple news sites.
• The personalized blocking mechanism reaches 89% accuracy by learning from user preferences and blocking similar headlines in future visits.
• The browser extension successfully warns users about potential clickbaits during real-time browsing.
• User-driven feedback significantly improves the system's ability to block unwanted content over time.
• The system performs consistently across diverse online news media platforms in both offline and online evaluation settings.
• The combination of automated detection and user personalization leads to a substantial reduction in exposure to misleading headlines.