Analysis of title influence on popularity of forum postings using text mining: Case study on Kaskus.com

The increasing number of internet user influence the rise of internet data. One of the increasing internet data takes the form of online forum postings. These forum postings could be accessed from numerous postings portals on the internet. Some portals show only the posting title to let the readers decide whether they want to read it or not by reading the title only. In this study, the author study whether an online postings headline or title could influence reading interest in the community by using the Naïve Bayes and decision tree methods as a comparison. Based on analysis using Naïve Bayes and Decision Tree algorithm, the result shows that title of postings can influence as high as 73% popularity of the postings.