The Analysis of Graduate Studies on Big-Data on Social Media through Text Mining

Abstract

Texts may contain useful information on many topics. Analyzing texts can help people make better decisions, do more effective work, and access more information. Data obtained from rich sources such as social media constitute big data belonging to these texts. Various methods are employed to understand and interpret these data. Among them, text mining and data analytics are the most widely used techniques. In addition, there may be need for more data than available through structured data to excavate the information contained in a given text data. This article examines graduate theses prepared in Turkey employing big data approach obtained from social media. These studies have been prepared by various departments and hence big data has been examined from various aspects. In this regard, thispaper provides brief summaries of some these theses. The findings reveal that the majority of related theses were written in the field of computer engineering. However, their characteristics differ from each other. While some target the software aspects, others analyze social media information. The next most popular field is the various departments in the field of communications. It has been observed that the number of theses written on big data has increased over the years. This study has applied word analysis on theses written between 2008 and 2022 through the text mining method. The results confirm the congruity of the word distribution in theses to the power law distribution. The overall findings point to the problem of excessive focus in theses.