Aggregating Research Papers from Publishers’ Systems to Support Text and Data Mining: Deliberate Lack of Interoperability or Not?

Knoth, Petr and Pontika, Nancy (2016). Aggregating Research Papers from Publishers’ Systems to Support Text and Data Mining: Deliberate Lack of Interoperability or Not? In: INTEROP2016 (Eckart de Castilho, Richard; Ananiadou, Sophia; Margoni, Thomas; Peters, Wim and Piperidis, Stelios eds.), 23 May 2016.

URL: http://interop2016.github.io/Program

Abstract

In the current technology dominated world, interoperability of systems managed by different organisations is an essential property enabling the provision of services at a global scale. In the Text and Data Mining field (TDM), interoperability of systems offering access to text corpora offers the opportunity of increasing the uptake and impact of TDM applications. The global corpus of all research papers, i.e. the collection of human knowledge so large no one can ever read in their lifetime, represents one of the most exciting opportunities for TDM. Although the Open Access movement, which has been advocating for free availability and reuse rights to TDM from research papers, has achieved some major successes on the legal front, the technical interoperability of systems offering free access to research papers continues to be a challenge. COnnecting REpositories (CORE) (Knoth and Zdrahal, 2012) aggregates the world’s open access full-text scientific manuscripts from repositories, journals and publisher systems. One of the main goals of CORE is to harmonise and pre-process these data to lower the barrier for TDM. In this paper, we report on the preliminary results of an interoperability survey of systems provided by journal publishers, both open access and toll access. This helps us to assess the current level of systems’ interoperability and suggest ways forward.