Topic Modeling and its Applications

Abstract

The amount of data online is growing rapidly, and while most people think about numbers when they hear data, there is a huge amount of textual data available that can prove to be useful. Analytics involving numbers has been pushed to extreme ends, and analyzing text has the capability of going just as far. I will detail the history, challenges, and applications of text analysis including topic modeling. This paper discusses the basic information and history of text analysis, and then focuses on one particular area: topic modeling. Topic modeling is a natural language processor and its goal is to determine topics that are related to documents and words within a corpus of text. After discussing my personal experience with topic modeling, I will consider examples of how to carry out topic modeling on a sample data set, and why each topic model is useful in terms of data analysis.