Clickbait Headline Detection Using Supervised Learning Method

Abstract

With the increasing use of the Internet of Things (IoT) as a means of communication in the 21st century, many news media now rely on the internet as an online news publication platform. News headlines are often made as attractive as possible to entice the reader’s curiosity and thus increase the views of a news article. One of the many tactics employed is the use of clickbait. This research involves creating a model to detect headlines that contain clickbait. The model can act as a classifier between real news and clickbait-filled headlines with a Natural Language Processing (NLP) method approach. Bidirectional Long Short-Term Memory (Bi-LSTM), Decision Tree, and K-Nearest Neighbor (KNN) are all methods that can be used to distinguish actual news headlines from clickbait-laden headlines. This work is preliminary as research in this field is still being conducted, and improvements to the accuracy of these systems are still improving.