<![CDATA[Undoubtly that social media has been used to exchange idea easier and faster. That increasingly widespread exchange of data on social media boosts the data growth and generates various structures of data. University can benefit this large of data by gaining knowledge related to topics discussed by social media users. By gaining this knowledge, university can define which information must be disseminated by analyzing the topic that most of users discussed on social media. However, the diversity of data structures spread on social media complicates the process of obtaining this knowledge. This can be overcome by analyzing social media data using the Natural Language Processing method. This study suggests a framework for text classification using topics in the form of Indonesia language that is spread on social media. The data analyzed in this research are social media data on Facebook and Twitter at the XYZ University. This analysis was carried out by conducting experiments using applications from previous research]]>
