This is a collection of the text processing and NLP implementations :
-
Term Frequency Word Cloud
- A script to parse any PDF file and output a word cloud of the 50 most frequent terms.
-
Bayesian Autocorrect
- A simple autocorrect based on Bayes Theorem inspired by Peter Norvig's Spell Checker to autocorrect an incorrectly spelt search term.
-
Corpus Build
- A script that extracts text and compiles a corpus from a Wikipedia dump file using gensim.
-
Reddit Sentiment Analysis
- This directory contains a web scraper that will scrape the names of the top 100 subreddits and a sentiment analyzer that will compute the sentiment of each subreddit from the polarity of the top 100 comments in the 'hot' section.