CrossCheck Pilot: un projet de base de données pour la détection des plagiats
CrossRef pilote un projet de mise en place d’une base de données des publications scientifiques pour la détection du plagiat. Les systèmes de détection de plagiat existants, qui fonctionnent par comparaison entre un manuscrit soumis et une base de contenus publiés, n’indexent pas la plus grande partie du contenu des bases accessibles sous abonnement, bases publiées, justement, par les principaux éditeurs scientifiques collaborant au sein de CrossRef pour le cross referencing. Détails sur le site de CrossRef et extraits ici après le saut…
Current state-of-the-art in PD works by comparing a submitted manuscript to a database of published content. When the system detects a degree of textual overlap between the submitted manuscript and items in the database, the system produces an originality report, which can be used by editorial staff in order to determine whether the overlap is benign, or a case of plagiarism. The problem with these PD systems is that they are only effective if the database of published content contains items that are relevant to the domain of the documents being checked. Without a database of relevant existing published scholarly and professional content, it is impossible to effectively check purportedly scholarly/professional new content for plagiarism.
Currently, existing PD systems do not index the majority of scholarly/professional content because it is inaccessible to crawlers directed at the open web. The only scholarly literature that is currently indexed by PD systems is that which is available openly (e.g. OA, Archived or illegitimately posted copies) or that which has been made available via third-party aggregators (e.g. ProQuest). This, in turn, means that any publisher who is interested in employing PD systems in their editorial work-flow is unable to do so effectively. Even if a particular publisher doesn’t have a problem with plagiarized manuscripts, they should have an interest in making sure that their own published content is not plagiarized or otherwise illegitimately copied.
needs to be indexed. Hence, the main ial goal of CrossCheck will be to facilitate the indexing of publisher full text content by PD services and to set common terms and conditions for the use of publisher full text by PD services. CrossCheck will also help facilitate the use of PD services by publishers and provide advice to publishers on best practices in using a PD service.
Classé dans:communication scientifique, publication et éditeurs | 2 Comments
Étiquettes : CrossCheck, CrossRef, plagiarism, plagiat