Design And Development Of Optimal Semantic Text Tokenization (OSTT) Method For Clickbait Pre-Processing

Abstract

Generally, the usage of social media is increasing. It leads to the increase in online advertisement to be more popular. But these advertisements accompanied with the disturbing clickbait headlines. It is spreading the headlines with irrelevant messages. But the users may get dissatisfaction because the content of the article doesn’t match with their expectation. So, it is an important task to prediction of clickbaits in social network to fight this problem. In order to click the fake link and attract the user’s attention the click bait uses good expression with good phrases It means that clickbait use false titles in order to obtain information about the hidden user from the target page. However, it is extremely difficult to foresee and recognize these headlines manually. As a result, there is a need to design an intelligent system for predicting clickbait in social networks. Before developing a method, pre-processing of text is playing key role to improve the prediction accuracy. This research work developed a new method to pre-process the text. There are four steps in pre-processing such as Tag and special characters removal, Tokenizing, stopwords removal, stemming and Lemmatization.