Heterogeneous Queries for Synoptic and Phrasal Search

Warning

This publication doesn't include Faculty of Arts. It includes Faculty of Informatics. Official publication website can be found on muni.cz.
Authors

SUCHOMEL Šimon BRANDEJS Michal

Year of publication 2014
Type Article in Proceedings
Conference CLEF2014 Working Notes
MU Faculty or unit

Faculty of Informatics

Citation
Web http://ceur-ws.org/Vol-1180/
Field Informatics
Keywords suspicious document; plagiarism detection; search engine; source retrieval; stop word; text alignment; snippet similarity;
Attached files
Description This paper describes our approaches for the Plagiarism Detection – Source Retrieval task of PAN 2014. We combined and improved methodology used at PAN 2012 and PAN 2013. Our system combines three types of queries: The keywords-based queries; the paragraph-based queries; and the headers-based queries. The queries are distinguished also by other properties such as the phrase query or the positional query. The queries are submitted to two search engines – Chatnoir and Indri – according to their properties. The query’s position serves for the search control, minimization of the total number of executed queries is the system’s priority. Downloaded documents are textually compared with the suspicious document and if a similarity is found, the downloaded document is reported.
Related projects:

You are running an old browser version. We recommend updating your browser to its latest version.