Building a ChatBot with Python, NLTK and scikit
summary
Walks through runnable Python notebooks that build a Q&A chatbot using a SQL-backed core, uses NLTK tokenization and POS tagging and the Stanford CoreNLP dependency parser (Java 8 required), extracts triple-based features from tagged sequences, trains a scikit-learn RandomForest on ~100 labeled sentences and reports about random-forest sentence classifier ~90% on small set, and deploys a simple TCP client/server on an Oracle Cloud VM; demos work but are flaky and the approach trades scale for interpretable, grammar-aware retrieval.
AI-written summary · about
chapters (8)
- 0:00Chatbots and NLP basics
- 11:57Python setup and notebooks
- 24:40Tokenizing with NLTK
- 35:33Stemming lemmatizing limits
- 36:38Parsing with Stanford NLP
- 50:13Sentence classification
- 1:14:49FAQ chatbot build
- 1:20:08Live demo
related talks
33mBuilding the next-generation Conversational AI with Python and Deep Learning
11mCreate a Custom Chat Bot with Python
01h 14mIntroduction to NLP
39mHOWTO: Teach Python to Read
39mAutomate the Boring Stuff with Slackbot
02h 19mWorkshop: Demystifying Natural Language Processing using Python (Scikit-Learn/ Keras)
01h 39mNatural Language Processing (NLP) in Python - From Zero to Hero
44mLightning Talks