CourseNatural Language Processing: The Complete Course · Module 5: Evaluation, Sentiment and Topics · part 36 of 73
Part 36 · Module 5: Evaluation, Sentiment and Topics

Lesson 4: Topic modelling: what is the inbox about?

45 min read·14 Oct 2026
Module 5 · Lesson 4 · About 50 minutes · Needs: basic Python; Module 4 Lesson 2 (TF-IDF) helps but is not required

The situation

It is Monday 5 October 2026, project week 5. Rosa wants a Monday report: what are customers writing about, and is any of it getting worse? Ines's seven intents do not answer that: warranty_fault lumps a snapped spoke with a battery that will not charge.

This lesson lets the 1,428 English messages (simulated) suggest the themes, names them, and counts one week by week. The running example is TW-M0733, from Arjun Sharma on Tuesday 9 June 2026, the day Tidewell emailed Volt 2 owners about battery batch VB-0426.

ROSA'S MONDAY QUESTION 1,428 English messages no labels topic model groups words that occur together themes counted per week for Rosa Is any theme getting worse? The counts by week answer it.
Rosa's report needs themes nobody has listed: the messages (blue, our data) go into a topic model (amber, the model's guess), and themes counted per week come out (green).

The rest of this part opens with an account

Signing up is free and takes one email. Nothing to pay, and nothing renews.