/UoA-NLP

Work done researching NLP & Machine Learning at the University of Auckland.

Primary LanguagePython

UoA-NLP

This is a summary on how to run the work done on this project and alter the parameters.

Dependencies

  • Python3
  • NLTK, don't forget to run nltk.download('popular') to install used popular resources
  • PostgreSQL and the python driver psycopgl
  • scikit-learn for Machine Learning
  • gensim for LDA/LSA
  • Optional:
    • IPython (IPyKernel, Jupyter) for LDA visualisation
    • pandas and redis for the recommender

The project needs a postgreSQL server running a specific database, which creation and population scripts are available on the VM under /data/uoa-xero-db/. The Configuration.py in the Database module should also be modified correctly.

Usage

Running the algorithms

the main scripts are located at the root of the repository. To run the classifier and clustering algorithm:

python break_classify.py [algorithm]

where [algorithm] is one of the implemented clustering algorithms. At the present time, it can be dbscan, hierarchical,

Customizing the parameters