Text Reuse @ UCREL & NLPT
Urdu Paraphrase Plagiarism Corpus (UPPC)
This corpus contains 160 Urdu text documents in total. 20 documents are original Wikipedia articles on well-known people whereas 140 documents (manually created by volunteers) are paraphrase plagiarise and non-plagiarise versions of the original articles. 75 documents are paraphrased by 5 university students using different paraphrasing techniques. 65 documents are independently written without considering the source article.
Download the corpus: UPPC.zip
Licence of the Corpus
The corpus is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Corpus Statistics
The corpus contains 20 original documents (Wikipedia excerpts), 75 plagiarised and 65 non-plagiarised documents of length between 200-300 words. Total number of words (tokens) in the corpus are 46,729 words (tokens), 6,201 unique words and 2,711 sentences.
Corpus Format
The corpus has been saved in standard XML format for easy access and use. All documents of the corpus are saved inside the "data" directory. Each document start with XML declaration tag. Corpus text has been inclosed within <UPPC_document> tag which contains 3 attributes; classification: Three possible values O (Original), P (Paraphrased), NP (Non-Paraphrased) title: Name of the personality on which the document is written file: Filename of the document
The "all_files.txt" file provided with the corpus contains, in CSV format, list of plagiarised and non-plagiarised file names and their relationship (P or NP) with the source(s).
Acknowledging the Corpus
We release this corpus with an intention that it will foster the research in Urdu paraphrase plagiarism detection systems. If you use the corpus we kindly ask you to refer to it in your publications as follows:
Bibtex:
@InProceedings{SHARJEEL16.364,
author = {Muhammad Sharjeel and Paul Rayson and Rao Muhammad Adeel Nawab},
title = {UPPC - Urdu Paraphrase Plagiarism Corpus},
booktitle = {Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC 2016)},
year = {2016},
month = {may},
date = {23-28},
location = {Portoroz, Slovenia},
editor = {Nicoletta Calzolari (Conference Chair) and Khalid Choukri and Thierry Declerck and Sara Goggi and Marko Grobelnik and Bente Maegaard and Joseph Mariani and Helene Mazo and Asuncion Moreno and Jan Odijk and Stelios Piperidis},
publisher = {European Language Resources Association (ELRA)},
address = {Paris, France},
isbn = {978-2-9517408-9-1},
language = {english}
}
The corpus also has a permanent DOI from Lancaster University: 10.17635/lancaster/researchdata/67
Contact Information
If you have any questions or suggestions, or any other feedback please send an email to ucrel@lancaster.ac.uk
Please see the README file included with the corpus for further information. Note: The use of this corpus is restricted to research and/or academic purposes only.
|
|
|