Thanks to visit codestin.com
Credit goes to github.com

Skip to content

rakshajalan/ECIR-2018

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 

Repository files navigation

ECIR-2018

Medical Forum Question Classification using Deep Learning

   Raksha Jalan, Manish Gupta, Vasudeva Varma

Following are folder names having respective files.

1)Dataset: It contains both training and test dataset in tsv format. Subfolder named "Crawled_questions_for_Weak _Supervision" contains text file having questions those re crawled seperately from "med-help.org" where title and corresponding question is seperated by sign "<==>".

2)Self_training:Contains codes corresponding to basic self-training algorithm and self-training with lookups. 3)Supervised: Contains codes for all supervised models mentioned in Table 2 of paper. 4)Weakly_Supervised:Contains codes for weakly supervised models mentioned in Table 3 of paper.

How to Run code? For Supervised models: Inputs are Training,Test DataSet and Glove vectors. For SoA based networks: Aditional input is output file of MetaMap.Where it contains all medical entities of respective questions along with their CUIs.

For Self_trainig Models: Inputs are Training,Test DataSet,Crawled Data and Glove,TF-IDF vectors. *Models stores tuned parameters those are learned from self training in pickle format.

For Weakly Supervised Models: Inputs are Training,Test DataSet,Pickle files to load the same parameters those are obtained from self-training Models.
Make changes in code to provide corresponding inputs and run the codes.

About

Medical Forum Question Classification using Deep Learning

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages