MatlabNLP is a collection of efficient algorithms, data structures and well-tested functions for doing Natural Language Processing in the Matlab environment. We are continuously speeding up the underlying algorithms and functions. You can use the whole library or you can just grab functions and methods that you need and copy them onto your project folders.

Objective

To produce a collection of useful NLP tools for MATLAB

Components

• Tokenizers
  • tf-idf
  • Term Frequency (multinomial featurizer)
  • Bernoulli Featurizer
• Stemmers
• Reverse Stemmer
• Useful Corpora
  • Declaration of Independence
  • English conversations
• Learning Models
  • Naive Bayes
  • Logistic Regression
  • SVM
  • Linear Regression
• Evaluation Measures
  • Precision
  • Recall
  • Fallout
  • F1-measure
  • MSE (mean square error)