Topic Discovery

The main goal of topic modeling is to find a set of topics that can be used to classify a set of documents. These topics are implicit because we do not know what they are beforehand, and they are unnamed. We just generally assume that some documents are similar to each other and that we can organize them into topics.

The number of topics is usually small; that is, from 2 to 10. However, there are some use cases in which you may want to have up to 100 (or even more) topics. Since it is the computer algorithm that discovers the topics, the number is generally arbitrary. These topics may not always directly correspond to topics a human would identify. In practice, the number of topics should be much smaller than the number of documents. This helps the topic modeling algorithm in the sorting process. The more examples of documents that we provide, the better the accuracy with which the algorithm can sort and place the documents into categories.

The number of topics...