A distributed chinese n-gram language model implementation for train and test on large corpus , using Hadoop MapReduce.

Word Prediction and Auto-correct using NLP tools (n-gram backoff, viterbi decoding as a spell checker, letter-level backoff prediction) in Java.

Analyzes text and gathers data into n-grams. Gives the top x n!-grams. Supports mySQL databases.

Assignment 2 of the course 'Distributed Systems Programming' by Meni Adler. In the assignment we build an application that calculates the probabilities for any word to come after a couple of words, for ANY couple of words in the n-gram corpus (google).

"Advanced Data Structures and Algorithms" Course project