A systematic and analytical approach to techniques and tools in topic modeling

Abstract

Topic modeling is one of the recent upcoming research areas of interest among the researchers. Topic Modeling is a straightforward way to examine the huge volumes of unstructured data. Each topic is a collection of words and these words usually bond together more frequently. When this technique is applied to a huge volume of data it can join words having same meanings and distinguish the uses of words with multiple meanings. The intention is to study and examine different topic modeling algorithms and to perform a brief literature review and analysis was performed and the obtained results are presented in this paper. Many techniques for topic modeling proposed by different researchers are put together and characteristics and drawbacks of various techniques have discussed. We present this paper with the intention that it will help few of the researchers in finding out the problems, present challenges and future scope of research in topic modeling.