Code for programming assignments in C++ and bash from the Coursera course Text Mining and Analytics, given by ChengXiang Zhai from the University of Illinois Urbana-Champaign
Run POS (Part-of-Speech) tagging on a text document. POS tagging refers to the process of assigning part-of-speech tags, such as “noun” or “verb,” to words in documents.
- task-1.sh - Script for task 1.
Mine the words with the strongest syntagmatic relations in a corpus of 10,000 restaurant reviews extracted from the Yelp Academic Dataset. We are going to focus on implementing Mutual Information as the association measure to mine the top K syntagmatically related words in the Yelp corpus.
- association.cpp - Implementation of Mutual Information function for syntagmatic relation discovery.
- task-2.sh - Script for task 2.
Implementtationof Probabilistic Latent Semantic Analysis (PLSA), which is the most basic topic model, and also one of the most useful ones for mining topics and themes from text collections. PLSA assumes that each document in the corpus is generated from a set of topics represented by a unigram language model. Every document is also associated with a set of mixing weights that determine the proportions of topics in the document. In addition to the topic models, a fixed background language model can be also used to explain the generation of very frequent words.
- config.toml - Configuration file for MeTA toolkit.
- plsa.cpp - Implementation of E-Step for Probabilistic Latent Semantic Analysis.
- task-3.sh - Script for task 3.
Use of restaurant reviews to predict whether a set of given restaurants will pass the public health inspection tests. We are going to train a classifier based on the first 546 restaurants (labeled), and then predict the labels of the last 200 restaurants.