This notebook explores how clustering semantically similar words can help make Natural Language Processing tasks easier.

Implemented a collection of Ngram language models on brown corpus from scratch

This project trains a Long Short Term Memory (LSTM) network to detect and classify a text written in English according to a particular variant: whether it is British or American.

Text Analysis techniques using Brown Corpus , CMU dictionary

A program which guesses next words based on the user's input. Suggestions are the words with the highest probability to follow what has been already written, calculated in the n_grams of different size.

Natural Language Processing (2018)

Fun in-class exercise for understanding the inner workings of word2vec in NLP. Implemented Google News 300 word2vec pre-trained model, and also trained a model from scratch with an existing text dataset (Brown Corpus).

POS tagging using a Hidden Markov Model (HMM) with Viterbi Decoding

Corpus Linguistics slides, labs, assignments and data

Quantify the similarity between pairs of words of a dataset using Lin similarity, NPMI and LSA.

Part-Of-Speech-tagging using Hidden Markov model to identify the category of words ('noun', 'verb', ...) in plain text.

Hidden Markov Model for Part of Speech Tagging

Simple Python Implementation of Stemmer and Lemmatizer

Sentence generator using tokens from the Brown corpus

Viterbi Algorithm for POS tagging of sentences using Brown corpus

Various exports from Brown Corpus and useful scripts.

Auto tagger created with RNN using Bi-LSTM cell