Skip to content

A Hybrid Approach to Answer Selection in Question Answering Systems

License

Notifications You must be signed in to change notification settings

duonghuuphuc/anselect17

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 

Repository files navigation

Answer Selection for Python (Sept, 2017)

Code for the paper A Hybrid Approach to Answer Selection in Question Answering Systems (IUKM, 2018)

Requirement

  • Source code is written in Python 3.5 (Anaconda Environment), and requires Keras (TensorFlow [GPU] back-end)
  • Use the pre-trained word2vec, published by Mikolov et al.

Run the model

  • Download all pre-processed datasets in data directory
  • Run this command for evaluating Feed-forward neural network model: python ann.py
  • Run this command for evaluating Convolutional neural network model: python conv1d.py
  • All above programs perform three phases: (1) build the model, (2) compute accuracy on training and testing datasets, and (3) save model to disk

Contributors

Releases

No releases published

Packages

No packages published