in this first session, you had studied the notion of ‘aboutness’- text is usually about something . For example, a document consisting of the terms ‘MIT’, Stanford, Cambridge’ is probably about education, the best universities, research ,or all of these ‘topics’. these topics may not be explicitly present in the document, but they might still be there.
Topic modelling is the art and science of identifying ‘latent topics’ in text. Let’s understand the basic of topic modelling.
You saw that topic modelling is an unsupervised technique used to model topics within documents. But how exactly do we define topics?
Let’s get some intuition of what it means to identify topics in a document.