End-to-end Representation Learning for Question Answering with Weak Supervision

Knowledge base Question Answering system for in the participation the QALD-7 Shared task

This is an accompanying repository for our 4th SemWebEval Challenge at ESWC 2017 paper (full-paper). It contains the code that was used to produce the results for the QALD-7 shared task.

This repository contains experimental software and is published for the sole purpose of giving additional background details on the respective publication.

Please use the following citation:

@inproceedings{TUD-CS-2017-0113,
	title = {End-to-end Representation Learning for Question Answering with Weak Supervision},
	author = {Sorokin, Daniil and Gurevych, Iryna},
	publisher = {Springer, Cham},
	series = {Communications in Computer and Information Science},
	volume = {769},
	booktitle = {Semantic Web Challenges: 4th SemWebEval Challenge at ESWC 2017},
	pages = {70-83},
	month = oct,
	year = {2017},
	location = {Portoroz, Slovenia},
	website  = {https://doi.org/10.1007/978-3-319-69146-6_7},
}

Paper abstract:

In this paper we present a knowledge base question answering system for participation in Task 4 of the QALD-7 shared task. Our system is an end-to-end neural architecture for constructing a structural semantic representation of a natural language question. We define semantic representations as graphs that are generated step-wise and can be translated into knowledge base queries to retrieve answers. We use a convolutional neural network (CNN) model to learn vector encodings for the questions and the semantic graphs and use it to select the best matching graph for the input question. We show on two different datasets that our system is able to successfully generalize to new data.

Please, refer to the paper for more the model description and training details

Contacts:

If you have any questions regarding the code, please, don't hesitate to contact the authors or report an issue.

• Daniil Sorokin, <lastname>@ukp.informatik.tu-darmstadt.de
• https://www.ukp.tu-darmstadt.de
• https://www.tu-darmstadt.de

Project structure:

File	Description
questionanswering/construction	Base classes for semantic graphs
questionanswering/datasets	Datasets IO
questionanswering/models	Model definition
questionanswering/wikidata	SPARQL query definitions for Wikidata
resources/	Necessary resources

Requirements:

• Python 3.6
• PyTorch 0.3.0 - read here about installation

Reproducing the shared task results:

Please contact the first author if you would like to reproduce the results or need additional information on the experiments. This repository only contains the main code for the neural model, the pre-trained models and the Wikidata knowledge base are not included.

License:

• Apache License Version 2.0