The Real Python Podcast Exploring Modern Sentiment Analysis Approaches in Python
55 snips
Dec 20, 2024 Jodie Burchell, a developer advocate for data science at JetBrains with a PhD in clinical psychology, shares her expertise on sentiment analysis. She discusses traditional lexicon-based methods and advanced machine learning techniques, highlighting the evolution of sentiment analysis with large language models. Jodie emphasizes the challenges of linguistic nuances and context in emotional classification. From practical applications in blogging to tools for analysis like TextBlob, she provides valuable insights for anyone looking to dive deeper into this field.
AI Snips
Chapters
Books
Transcript
Episode notes
Jodie's Conference Talks
- Jodie Burchell focused on LLMs and specific interests like hallucinations and LLM sentience.
- She keynoted at PyCon Italia and other conferences, discussing LLM performance measurement.
Jodie's Background
- Jodie's PhD was in psychology, focusing on emotions and relationships, especially hurt feelings.
- This led her to sentiment analysis as a way to automatically detect emotions in text.
Dimensional Emotion
- Dimensional emotion classification goes beyond simple positive/negative.
- It considers polarity, arousal (physiological response), and dominance (expression).
