CW Corpus

Handle:	https://hdl.handle.net/21.11129/0000-000B-D363-7 (persistent URL to this page)

The Complex Word (CW) Corpus contains 731 sentences each with one annotated CW. These simplifications were mined from Simple Wikipedia edit histories. Each entry gives an example of a sentence requiring simplification by means of a single lexical edit. This resource is primarily designed for the evaluation of CW identification systems.

Download

DistributionLicence

CC - BY - SA

Restrictions: Academic - Non Commercial Use, Attribution, Share Alike

User Nature: Academic

Attribution Details: Shardlow, M. (To appear). The CW Corpus: A new resource for evaluating the Identification of Complex Words. In Proceedings of the Second Workshop on Predicting and Improving Text Readability for Target Reader Populations (PITR 2013), Sofia, Bulgaria, Association for Computational Linguistics

Distribution Access/Medium: Accessible Through Interface

Licensors:

Matthew Shardlow Male

University of Manchester

[javascript protected email address]

School of Computer Science

[javascript protected email address]

Distribution rights holders:

Matthew Shardlow Male

University of Manchester

[javascript protected email address]

School of Computer Science

[javascript protected email address]

IPR Holder

Matthew Shardlow Male

University of Manchester

[javascript protected email address]

School of Computer Science

[javascript protected email address]

Contact Person

Matthew Shardlow Male

University of Manchester

[javascript protected email address]

School of Computer Science

[javascript protected email address]

text

Monolingual text corpusLanguages

English

Linguality

Linguality type: Monolingual

Size

731 Sentences

Character encoding

UTF - 8

Domains

General

Modalities

Written Language

AnnotationOther

Segmentation level: Sentence, Word

Format: Text

Annotation Mode: Manual

Interannotator Agreement: 97.5% Kappa

Creation

Creation mode: Manual

Original Sources

Simple Wikipedia

Resource Creation

Resource Creator

Matthew Shardlow Male

University of Manchester

[javascript protected email address]

School of Computer Science

[javascript protected email address]

Metadata

Created: 06/12/2013

Last Updated: 06/12/2013

Metadata Creator

Matthew Shardlow Male

University of Manchester

[javascript protected email address]

School of Computer Science

[javascript protected email address]

Usage

Foreseen UseNlp Applications

Use NLP Specific: Natural Language Understanding, Text Generation

Documentation

Tool Documentation: Online

Document Type: Other

Matthew Shardlow, The CW Corpus, http://www.nactem.ac.uk/meta-net/Narratives/CW-Corpus.pdf , 2013

People who looked at this resource also viewed the following:

People who downloaded this resource also downloaded the following: