Text Reuse with ACL: (Upward) Trends

Proceedings of the ACL-2012 Special Workshop on Rediscovering 50 Years of Discoveries, pages 76–82, Jeju, Republic of Korea, 10 July 2012. c⃝2012 Association for Computational Linguistics Text Reuse with ACL: (Upward) Trends Parth Gupta and Paolo Rosso Natural Language Engineering Lab - ELiRF Department of Information Systems and Computation Universidad Polit´ecnica de Valencia, Spain http://www.dsic.upv.es/grupos/nle {pgupta,prosso}@dsic.upv.es Abstract With rapidly increasing community, a plethora of conferences related to Natural Language Processing and easy access to their proceed- ings make it essential to check the integrity and novelty of the new submissions. This study aims to investigate the trends of text reuse in the ACL submissions, if any. We car- ried a set of analyses on two spans of five years papers (the past and the present) of ACL using a publicly available text reuse detection appli- cation to notice the behaviour. In our study, we found some strong reuse cases which can be an indicator to establish a clear policy to handle text reuse for the upcoming editions of ACL. The results are anonymised. 1 Introduction Text reuse refers to using the original text again in a different work. The text reuse in its most general form can be of two types: verbatim (quotations, defi- nitions) and modified (paraphrasing, boilerplate text, translation). Although, the text reuse can be legal or illegal from a publishing authority perspective about the accreditation to the original author, more impor- tantly it involves the ethical issues, especially in the scientific work. There is a fuzzy line between the text reuse and the plagiarism and often this line is legislative. There are no straight-forward measures to declare a work as plagiarism and hence the publishing houses usually deploy their own rules and definitions to deal with plagiarism. For example, IEEE1 and ACM2 both consider the reuse as plagiarism in case of: 1. unaccredited reuse of text; 2. accredited large portion of text without proper delineation or quotes to the complete reused portion. IEEE does not allow reusing large portion of own previous work, generally referred as self reuse or self plagiarism, without delineation, while ACM al- lows it provided the original source being explicitly cited. With the advent of a large number of conferences and their publicly available proceedings, it is ex- tremely easy to access the information on the desired topic to refer to and to reuse. Therefore, it becomes essential to check the authenticity and the novelty of the submitted text before the acceptance. It becomes nearly impossible for a human judge (reviewer) to discover the source of the submitted work, if any, unless the source is already known. Automatic pla- giarism detection applications identify such poten- tial sources for the submitted work and based on it a human judge can easily take the decision. Unaccredited text reuse is often referred to as plagiarism and there has been abundant research about the same (Bouville, 2008; Loui, 2002; Mad- dox, 1995). Self plagiarism is another related is- sue, which is less known but not less unethical. …

Content truncated to 500 words