CornellDataScience/QuACC

Name: QuACC

Owner: Cornell Data Science

Description: QuACC : Question Answering for Cornell Courses ?

Created: 2018-03-03 18:46:48.0

Updated: 2018-03-19 23:21:56.0

Pushed: 2018-03-19 23:21:55.0

Homepage:

Size: 15086

Language: Python

GitHub Committers

UserMost Recent Commit# Commits

Other Committers

UserEmailMost Recent Commit# Commits

README

Transfer Learning for QA Systems

Members: Kenta Takatsu (CS '19), Yuji Akimoto (CS '19), Chetan Velivela (CS '19)

Objective: To create a question answering system that can synthesize information from online resources such as textbooks and syllabi, and respond to questions about those resources. In the long term, we view this as an extremely useful tool for Cornell students that automatically answers easier questions asked on Piazza. In the short term, we will focus on creating a system to answer questions from the SQuAD dataset.

Sample QA

Model Architecture

R-Net


This work is supported by the National Institutes of Health's National Center for Advancing Translational Sciences, Grant Number U24TR002306. This work is solely the responsibility of the creators and does not necessarily represent the official views of the National Institutes of Health.