# Topic Modeling Algorithms

Topic modeling algorithms operate on the following assumptions:

- Topics contain a set of words
- Documents are made up of a set of topics

Topics are not observed but are assumed to be hidden generators of words. After these assumptions, different algorithms diverge in how they go about discovering topics. In this chapter, we will cover two topic modeling algorithms, namely **LSA** and **LDA**. Both models will be discussed in detail in the coming sections.

## Latent Semantic Analysis

We will start by looking at LSA. LSA actually predates the **World Wide Web**. It was first described in 1988. LSA is also known by an alternative acronym, **Latent Semantic Indexing** (**LSI**), particularly when it is used for semantic searches of document indexes. The goal of LSA is to uncover the latent topics that underlie documents and words. The assumption is that these latent topics drive the distribution of words in the document. In the next section, we will learn about...