search

cvlavanya / hw2-team17

Repository for Team Project
0 stars 0 forks source link

Enhancements #11

Closed cvlavanya closed 11 years ago

cvlavanya commented 11 years ago

To enhance the keyterm extraction task.

Currently, it can identify only the gene and protein names with certain accuracy, I plan to enhance it to include the other important words in the question (eg. "role","mutation" etc.) that would in turn increase the precisions of the document and passage retrieval tasks. Initially, I would include stemming and stop word removal. I plan to use the Lucene library for doing this task.

cvlavanya commented 11 years ago

Incorporated the discovery of verb forms in addition to gene names. Used Stanford CoreNLP toolkit and POS tagging.