Mining Text in News Channels: A Case Study from Facebook

Said Salloum, Mostafa Al-Emran, Khaled Shaalan


Recently, the usage of social media websites has become an attractive phenomenon in our daily life. These sites allow their users to communicate with each other through various tools. This results in learning and sharing of valuable information among their users. The nature of such information is categorized as unstructured and fuzzy text. The present study aims at analyzing textual data from Facebook and attempts to find interesting knowledge from such data and represent that knowledge in different forms. 33815 posts from 16 news channels pages over Facebook were extracted and analyzed. Findings revealed that there is a strong relationship between the Guardian and the Independent online news channels. Results indicated that there are four clusters in the study. Moreover, results showed that the overall collected data concentrated on three main topics: "Rio de Janeiro", "USA elections", and "UK leaves the European Union". These three main topics are considered as the hot topics that were discussed across all news channels provided by Facebook posts. Moreover, results depicted how the Text Parsing node can be employed to recognize terms and their examples in
the dataset that involves the text. Other implications and future work are presented in the study.

Full Text: