The data-literature interlinking service

Burton, Adrian; Koers, Hylke; Manghi, Paolo; Bruzzo, Sandro La; Aryani, Amir; Diepenbroek, Michael; Schindler, Uwe

doi:10.1108/prog-06-2016-0048

Cited by 20 publications

(7 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is the limitation of the Data Literature Interlinking database (Burton et al , 2017), to which the RDA workgroup contributes, which lists over 20,000 data sets and over 3,500 publications when queried for the CPS. Unfortunately, it is not clear what those items actually are.…”

Section: Literature Reviewmentioning

confidence: 99%

Framework for entity extraction with verification: application to inference of data set usage in research publications

Nestorov

Bačić

Jukić

et al. 2022

View full text Add to dashboard Cite

Purpose The purpose of this paper is to propose an extensible framework for extracting data set usage from research articles. Design/methodology/approach The framework uses a training set of manually labeled examples to identify word features surrounding data set usage references. Using the word features and general entity identifiers, candidate data sets are extracted and scored separately at the sentence and document levels. Finally, the extracted data set references can be verified by the authors using a web-based verification module. Findings This paper successfully addresses a significant gap in entity extraction literature by focusing on data set extraction. In the process, this paper: identified an entity-extraction scenario with specific characteristics that enable a multiphase approach, including a feasible author-verification step; defined the search space for word feature identification; defined scoring functions for sentences and documents; and designed a simple web-based author verification step. The framework is successfully tested on 178 articles authored by researchers from a large research organization. Originality/value Whereas previous approaches focused on completely automated large-scale entity recognition from text snippets, the proposed framework is designed for a longer, high-quality text, such as a research publication. The framework includes a verification module that enables the request validation of the discovered entities by the authors of the research publications. This module shares some similarities with general crowdsourcing approaches, but the target scenario increases the likelihood of meaningful author participation.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Framework for entity extraction with verification: application to inference of data set usage in research publications

Nestorov

Bačić

Jukić

et al. 2022

View full text Add to dashboard Cite

show abstract

“…A major step forward to solve this puzzle happened when publishers and data repositories joined forces through the Research Data Alliance and World Data System Working Group on the Scholarly Link Exchange (Scholix) iniative. Both publishers and repositories started contributing articledataset links to central Scholix hubs, like for example DataCite, CrossRef and OpenAire [13,14,15]. In this article we have investigated how much benefit research institutions have from combining Scholix information (article-dataset linking) with author and affiliation information (article-author and article-affiliation) from Scopus.…”

Section: Fig 1 Datacite Doi Registrations [6]mentioning

confidence: 99%

Why is getting credit for your data so hard?

et al. 2020

View full text Add to dashboard Cite

Institutions, funding bodies, and national research organizations are pushing for more data sharing and FAIR data. Institutions typically implement data policies, frequently supported by an institutional data repository. Funders typically mandate data sharing. So where does this leave the researcher? How can researchers benefit from doing the additional work to share their data? In order to make sure that researchers and institutions get credit for sharing their data, the data needs to be tracked and attributed first. In this paper we investigated where the research data ended up for 11 research institutions, and how this data is currently tracked and attributed. Furthermore, we also analysed the gap between the research data that is currently in institutional repositories, and where their researchers truly share their data. We found that 10 out of 11 institutions have most of their public research data hosted outside of their own institution. Combined, they have 12% of their institutional research data published in the institutional data repositories. According to our data, the typical institution had 5% of their research data (median) published in the institutional repository, but there were 4 universities for which it was 10% or higher. By combining existing data-to-article graphs with existing article-to- researcher and article-to-institution graphs it becomes possible to increase tracking of public research data and therefore the visibility of researchers sharing their data typically by 17x. The tracking algorithm that was used to perform analysis and report on potential improvements has subsequently been implemented as a standard method in the Mendeley Data Monitor product. The improvement is most likely an under-estimate because, while the recall for datasets in institutional repositories is 100%, that is not the case for datasets published outside the institutions, so there are even more datasets still to be discovered.

show abstract

“…There is currently no comprehensive aggregator for relationship information that could serve as this kind of general notification service. In the specific case of data-to-literature linking, the Scholix framework and associated Data-Literature Interlinking (DLI) Service have promise to serve key roles as a centralized third-party aggregator of relationship declarations (Burton, et al, 2017a;, but these initiatives are still in early days of building adoption.…”

Section: Requirements For Scholarly Resource Linkingmentioning

confidence: 99%

Scholarly resource linking: Building out a “relationship life cycle”

Mayernik

2018

Proc. Assoc. Info. Sci. Tech.

View full text Add to dashboard Cite

Scholarly resources, including publications, software, data sets, and instruments, are created in an iterative and interrelated fashion. Managing the relationships that exist among and between such resources is a central requirement for information systems. Practically, however, many scholarly resources exist online as discrete entities, divorced from other resources to which they are intimately related. A robust system for linking scholarly resources in a broad and sustainable fashion will have to navigate a set of complex and interrelated requirements. This paper presents results and insights from three different projects that focused on supporting more robust linkages among scholarly resources. The discussion details key technical and institutional challenges looking forward and backward in time across what might be considered to be a “relationship life cycle”: identifying, validating, characterizing, and preserving relationships. The goal of the paper is to help guide new research initiatives and operational services focused on integrating relationship information into the scholarly record.

show abstract

The data-literature interlinking service

Cited by 20 publications

References 15 publications

Framework for entity extraction with verification: application to inference of data set usage in research publications

Framework for entity extraction with verification: application to inference of data set usage in research publications

Why is getting credit for your data so hard?

Scholarly resource linking: Building out a “relationship life cycle”

Contact Info

Product

Resources

About