Learn Topic Modeling with LDA and LSA Models

Data with valuable insights can come in many forms, including unlabeled text data. In many practical cases, working with text data leaves us with massive amounts that are most likely unlabeled, messy, and difficult to decipher. In natural language processing, topic modeling is an unsupervised statistical method for extracting abstract topics from a collection of documents. 

In this project, we will follow step-by-step tutorials and explanations on data collection, data conversion, data cleaning, and topic model implementation. We will use the Federal Open Market Committee resource materials as our dataset to demonstrate how real-world data can be much more difficult than other curated datasets. We will learn how to implement topic models such as latent Dirichlet allocation (LDA) and latent semantic analysis (LSA) using Python libraries like gensim. This project will equip us with the basic knowledge and skills to collect, clean, and tackle almost any text data of interest.