What are current approaches for analyzing the emotions within a piece of text? What tools and Python packages should you use for sentiment analysis? This week on the show, Jodie Burchell, developer advocate for data science at JetBrains, returns to discuss modern sentiment analysis in Python.
👉 Links from the show: https://realpython.com/podcasts/rpp/232/
Jodie has a PhD in clinical psychology. We discuss how her interest in studying emotions has continued across her career.
Jodie covers three ways to approach sentiment analysis. We start by discussing traditional lexicon-based and machine-learning approaches. We then dive into how specific types of LLMs can be used for the task. We also share multiple resources so you can continue to explore sentiment analysis yourself.
This week's episode is brought to you by Sentry.
Topics:
- 00:00:00 -- Introduction
- 00:02:31 -- Conference talks in 2024
- 00:04:23 -- Background on sentiment analysis and studying feelings
- 00:07:09 -- What led you to study emotions?
- 00:08:57 -- Dimensional emotion classification
- 00:10:42 -- Different types of sentiment analysis
- 00:14:28 -- Lexicon-based approaches
- 00:17:50 -- VADER - Valence Aware Dictionary and sEntiment Reasoner
- 00:19:41 -- TextBlob and subjectivity scoring
- 00:21:48 -- Sponsor: Sentry
- 00:22:52 -- Measuring sentiment of New Year resolutions
- 00:27:28 -- Lexicon-based approaches links for experimenting
- 00:28:35 -- Multiple language support in lexicon-based packages
- 00:35:23 -- Machine learning techniques
- 00:39:20 -- Tools for this approach
- 00:42:54 -- Video Course Spotlight
- 00:44:15 -- Advantages to the machine learning models approach
- 00:45:55 -- Large language model approach
- 00:48:44 -- Encoder vs decoder models
- 00:52:09 -- Comparing the concept of fine tuning
- 00:56:49 -- Is this a recent development?
- 00:58:08 -- Ways to practice with these techniques
- 01:00:10 -- Do you find this to be a promising approach?
- 01:07:45 -- Resources to practice with all the techniques
- 01:11:06 -- Upcoming conference talks
- 01:11:56 -- Thanks and goodbye
👉 Links from the show: https://realpython.com/podcasts/rpp/232/