GitHub - ianleaman/PageRankKeywordClassifier: Exploring the use of Page Rank to extract key terms from a collection of documents.

ianleaman / PageRankKeywordClassifier Public

Notifications You must be signed in to change notification settings
Fork 0
Star 0

Exploring the use of Page Rank to extract key terms from a collection of documents.

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
friedman_articles		friedman_articles
.gitignore		.gitignore
README		README
results.txt		results.txt
run.py		run.py
simple_stopwords		simple_stopwords

Repository files navigation

This project explores the use of Google Page Rank as a means to generate key words from an input of simlar documents. 

It is a work in progress and was built to learn about Page Rank, not to create a better keyword classifier.

Currently the algorithm performs poorly compared to conventional keyword generation such as TF-IDF. Generally ranking stop words such as "the" or "it" highest.

Some improvements:
    - Expand to Ngrams or part of speach tags
    - Add the ability to give command line input.

To run on sample data:
    - Install Numpy
    - in terminal type "python run.py"