Workshop: Demystifying Natural Language Processing using Python (Scikit-Learn/ Keras)
summary
Hands-on workshop that actually builds and evaluates text models: uses Python with scikit-learn tools (count vectorizer, tf‑idf), trains a multinomial Naive Bayes SMS spam classifier on ~5,572 messages (4,825 ham / 747 spam) and reports ~98% accuracy, runs KNN examples and LDA topic modeling that narrows ten years of Reuters articles to ~25 clusters. Includes a Google Duplex demo to show limits of role‑based vs learned systems and a GitHub notebook with runnable code and worked outputs.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 0:52Who am I
- 2:16Pause
- 13:00Google Duplex
- 18:20Text Classification
- 24:30ACCME
- 25:23Text Summarization
- 27:00Machine Translation
- 29:53Routers
- 35:08Intuition
- 45:23Complex Questions
- 45:48Is NLP new
- 47:54Workshop
- 48:11Troubleshooting
- 49:06Github Page
- 50:16LD Avis
- 51:43Workshop Start
- 54:46What is Scikitlearn
related talks
01h 24mMachine Learning with Text in scikit learn
48mIntroduction to Natural Language Processing with Python
30mKick-Start your Understanding of Machine Learning with Python
31mDeep Learning Techniques for Natural Language Processing
44mTeaching Computers ABCs: A Quick Intro to Natural Language Processing
01h 38mNavigating the World of NLP and Text Analytics
01h 39mNatural Language Processing (NLP) in Python - From Zero to Hero
01h 27mTips, Tricks and Topics in Text Analysis