COREC – A neural multi-label COmmonsense RElation Classification system (ICPSR doi:10.11588/data/E5EHBV)

View:

Part 1: Document Description
Part 2: Study Description
Part 5: Other Study-Related Materials
Entire Codebook

Document Description

Citation

Title:

COREC – A neural multi-label COmmonsense RElation Classification system

Identification Number:

doi:10.11588/data/E5EHBV

Distributor:

heiDATA

Date of Distribution:

2019-10-22

Version:

1

Bibliographic Citation:

Becker, Maria, 2019, "COREC – A neural multi-label COmmonsense RElation Classification system", https://doi.org/10.11588/data/E5EHBV, heiDATA, V1

Study Description

Citation

Title:

COREC – A neural multi-label COmmonsense RElation Classification system

Identification Number:

doi:10.11588/data/E5EHBV

Authoring Entity:

Becker, Maria (Department of Computational Linguistics, Heidelberg University, Germany)

Date of Production:

2019

Distributor:

heiDATA

Date of Distribution:

2019-10-22

Study Scope

Keywords:

Arts and Humanities, Computer and Information Science, Semantic relation classification, ConceptNet

Topic Classification:

Commonsense relation classification, evaluation

Abstract:

<p>We examine the learnability of Commonsense knowledge relations as represented in CONCEPTNET. We develop a neural open world multi-label classification system that focuses on the evaluation of classification accuracy for individual relations. Based on an in-depth study of the specific properties of the CONCEPTNET resource such as relation ambiguity or argument heterogeneity, we investigate the impact of different relation representations and model variations. Our analysis reveals that the complexity of argument types and relation ambiguity are the most important challenges to address. We design a customized evaluation method to address the incompleteness of the resource that can be expanded in future work.</p>

Kind of Data:

program source code, python scripts

Methodology and Processing

Other Study-Related Materials

Label:

corec-master.zip

Notes:

application/zip