NLP — Natural Language Processing
“If 70 % of Earth is covered with Water, 70 % of Human Data is covered with Text”
2020 Notes
Spacy Cheat Sheet (link) — Use Wikipedia Pkg — Check Slideshare — Random Poem Generator (link)
Entity Recognition (link)
Reference doc (link)
Topics to be covered
Packages (NLTK, Text Blob, Spacy, urllib2, Beautiful Soap, pandas, numpy, matplotlib)
Basics of Installation Packages
Analytics Vidhya notes (Text blob, Intro to NLTK, Random Poetry Generator)
Theory (Tutorials Point)
Google Drive link
Apps to be created in Wolfram Language (Sentiment Analysis/Topic Classification)
Basic Functions
Applications (Chatbots/Sentiment Analysis/Topic Classification/Fake News Identifier)
APIs (Monkey Learn, AYLIEN)
Examples (Cortana, Siri — WolframAlpha API, Google Sheets Explorer, Watson)
Case Studies: Universe AI Text Generator / Watson AI beat World Champions in Jeopardy
Tidy Text in R (link) R Book
Python Book (link)
Machine Learning Example (Spam or Not, Fake or Not)
Markov Chain Analysis
Zipf’s Law
Data Sets to be Explored
Kaggle Sets (Fun with Simpsons)
Online Resources (Colab Research/Azure Notebooks/Trinket.io/Chatbots — Pandora Bots)
Generate Random Words
Data Structures
- String
- Corpus
- Document-term Matrix
Tokenization
Word Tokenize
WordPunct Tokenize
Unigrams, Bigrams and Trigrams
Stemming
Lemmatization
CheatSheets (NLTK — link + Text Blob Documentation (Link) + Spacy Cheat Sheet (link) + urllib2 (link) + Beautiful Soap (cheat sheet))
Twitter Sentiment Analysis (Workable Code (link))
Discussion on Wolfram Text Mining Techniques (link)
Real Time Text Analytics (link)
Monkey Learn API Integration with Google Sheets
Stanford NLP Page (link)
Online tools (link — Text analyser + Voyant Tools (link) ~ Entire Bird’s Eye View + Text Analysis online using Python (link) + link — Unigram/Bigram/Trigram Demo)
Azure Cognitive Services (link)
Short Challenges — Guess the Gender / Password Generator/Random Poetry Generator/ Hashtag Generator)
Topics to be covered
NLTK
Text Blob
Spacy (demo in it’s page) — Spacy.io
Games
Hangman Game
Guess the Animal
Wolfram School Projects
