Using Scattertext and the Python NLP Ecosystem for Text Visualization
summary
Open-source Python code produces interactive D3 charts embeddable in Jupyter using Scattertext with spaCy, Empath and agefromname and the demo corpus of 2012 convention speeches; it claims a novel scaled F score that privileges mid-frequency discriminative terms, uses PMI to keep some multiword expressions, and contrasts classes (e.g. Democrats vs Republicans) to reveal surprising patterns like job creators versus workers language. The package is pip-installable and the visuals work live in the notebook.
AI-written summary · about
chapters (12)
- 0:00Scattertext and NLP visualization
- 8:26Installing Scattertext
- 10:24Loading 2012 convention data
- 11:06Scaled F-score for term association
- 14:37Term frequencies and stop words
- 16:44Label overlap with tie-breaking
- 18:28Comparing term-finding algorithms
- 21:00Gender inference with agefromname
- 24:08Gender and party in language
- 25:59Topic and emotion modeling with empath
- 26:40Word vectors and deep learning
- 29:36Q&A
related talks
30mWords in Space
01h 51mNatural Language Processing in Python
36mData-visualisation with Python and Javascript: crafting a data-viz toolchain for the web
34mVisualising NLP pipelines with Pynorama
49mWhy You Should Do Text Analysis in Python (Even if You Don't Want to)
01h 38mEffective Data Visualization
01h 31mModern NLP in Python
24mLightning Talks