In the task of Topic Modeling $5$ we try to find some underlying semantic structure in a dataset using statistical methods. For example, we can cluster documents into topics with unsupervised methods. This tool can be used for data$-$analysis and data mining, with applications in NLP$,$ Bio$-$informatics, Software Engineering, Social networks and so on$.$ Some popular and traditional algorithms for this task are based on TF$-$IDF $($learned in Section $5).$