emorynlp / friendsqa

Question answering on multiparty dialogue

NLP4J

The NLP4J project provides a NLP toolkit for JVM languages. This project is under the Apache 2 license and is currently developed by the NLP Research Group at Emory University. Please join our forum to get notifications about new releases and give your feedback about this project.

• Latest release: 1.1.2 (06/29/2016).
• Release notes.
• Google groups.

Quick Start

• How to install.
• How to train.
• How to decode.
• NLP4J demo.

Components

• Tokenization.
• Morphological analysis.
• Part-of-speech tagging.
• Named entity recognition.
• Dependency parsing.
• Semantic role labeling (coming soon).
• Sentiment analysis (coming soon).
• Word2Vec & Struct2Vec (coming soon).

Supplements

• English lexica and models (hosted in bitbucket).
• Data format.

friendsqa's Issues

HOW TO RUN uop.py

All files in processors folder just consist functions but don't have programing running entrance(like name=main).how can I run this files?

Dataset Link goes 404

https://codeload.github.com/emorynlp/reading-comprehension/tar.gz/reading-comprehension-2.0 throws a 404 not found error. Please update the link in README.md

seems something wrong with code 'run_friends_split_context.py' and 'friendsqa.py'

when i try to run the code 'run_friends_split_context.py', it can not build the model from pretrain. After i fixed this bug, another bug cames.

It seems there are many tiny bugs in this code! would you fix it?

thanks

Correct way for data split

Hi.
I want to run some codes with FriendsQA dataset and find out that in the JSON file episode 21-22 are the test set and those after 23 are the development set, different from that written in README.

So which one is the correct way to split the dataset? Thanks.