This is a project in which I implemented two algorithms that disambiguate words given a sentence. The two implemented algorithms are:
- Personalised Pagerank as seen in this paper.
- Rudimentary wordnet similarity measures between words in a sentence.
I have also implemented and included Senseval 2 & 3 tests for both of these implementations, with detailed F1 scores generated for both.
This project requires python3.x in order to run, here are some required modules:
- nltk
- numpy
- scipy
- sklearn