Archive for the category: Open Data

Ms Ingrid Mason¹

¹AARNet, Canberra, Australia

DESCRIPTION

Linked open data techniques are being used to support resource discovery (of researchers and research objects) or for capturing research data.  The intent is to gather together individuals both expert and would-be expert so that interest in the development of a community of practice can be explored.  The desired outcome would be a list of community members that would like to share information, practices and occasionally meet online, to discuss their work (successes or challenges).

This BoF session is an open call to eResearch community members who are interested in discussing linked open data techniques in service of eResearch. Participants will be conversant with metadata standards e.g. Dublin Core, Darwin Core, RIF-CS, PROV, research data management practices. The format of the BoF will be an open discussion and the session 30 minutes in duration.

Biography

Ingrid Mason, Deployment Strategist with AARNet, provides support for engagement and the uptake of the national research and education network (NREN) and services with AARNet members across the research, cultural and collections sectors. Ingrid has worked on several NCRIS programs: Australian National Data Service, National eResearch Collaborative Tools and Resources, and Research Data Services.

http://orcid.org/0000-0002-0658-6095

Dr Mark Hahnel¹

¹Figshare, London, United Kingdom

ABSTRACT

In 2016, Figshare released the results of its global survey of 2,000 researchers in a report that assesses the global landscape around open data and sharing practices. The report highlights the extent of awareness around open data, the incentives around its use, and perspectives researchers have about making their own research data open. Some of the key findings of the 2016 report included:

• For the majority of respondents, open data is already a reality:
  • Approximately three quarters of respondents have made their research data openly available at some point; a similar number are aware of data sets that are open to access, reuse, repurpose and redistribute.
• Researchers place value on the credit they receive for making data open:
  • Nearly 70% of researchers value a data citation as much as an article citation. A further 10% value a data citation more than an article citation.
• Awareness of open data transcends age and career progression:
  • Encouragingly, Principal Investigators (PIs) and Professors consistently responded similarly to PhD students and Post-doctoral fellows in their awareness of open data usage.
• Respondents admit to uncertainty and gaps in their knowledge and are hungry for more information, perhaps one set of critical factors that hold back progression in open data sharing.
• Researchers are uncertain of how to cite datasets
• Researchers who have never made data openly available are considering doing so.

In 2017 figshare is partnering with Springer Nature, Wiley and Amazon Web Services to conduct the largest global study on researcher attitudes to open data. This talk will be a preview of the results which are due to be released during open access week (w/c 23rd of October 2017). Figshare have also been using these studies to help shape their data management and dissemination platforms

and engagement programs for publishers and institutions. The talk will conclude with a short demonstration of how we are incorporating feedback from the community to help foster a healthy and sustainable attitude to open science and research.

Biography

Dr. Mark Hahnel is the Founder of Figshare, a web-based platform that opens up scientific data to the world, making it available to anyone. He completed his PhD in stem cell biology at Imperial College London, having previously studied genetics in both Newcastle and Leeds. The idea for Figshare was first born in 2010 while Mark was studying at Imperial College London, before launching in 2012 as a portfolio company of Digital Science. He is a passionate and prominent member of the open science movement and its potential to revolutionise the research community.

Dr Amir Aryani¹ 

¹ANDS, Malvern, Australia

ABSTRACT

In this presentation, we report on an interoperability project between VIVO and Research Graph. Research Graph holds a large-scale network of scholarly records including publications, grants, datasets and researcher records. One of the capabilities of Research Graph network is the rapid integration with external repositories using XML crosswalks. The aim of this presentation is to demonstrate how to create a Research Graph database from an open access repository, enriching this graph  using  the international network of scholarly communication, and finally transform this graph database  to a VIVO  instance. Furthermore, we demonstrate this data can be transformed into the collaboration network visualisation in VIVO.

ABOUT VIVO AND RESEARCH GRAPH

Research Graph (researchgraph.org) is an open collaborative project derived from the outcome of the Research Data Alliance (RDA) working group on the Data Description Registry Interoperability [1]. This project addresses the problem of connecting  scholarly  works  across heterogeneous systems. The RDA working group recommendation provided a solution for connecting  publications  and  research  data  (data  in  research) across   multiple  open  access  repositories  using  co-authorship model and jointly funded research projects. The group had participants from Australian National Data Service (ANDS), Dryad (US),  CERN  InspireHEP  (Switzerland),  figshare  (UK),  da|ra  and GESIS (Germany), Data Curation Unit (Greece), OpenAIRE (European Infrastructure), ORCID, and DataCite.  Research Graph adopts and extends this work by creating a distributed graph that connects open access repositories to close research management systems traditionally locked behind the firewall [2].   In addition, the distributed  graph  addresses the challenge of scalability and enables individual universities and repositories to hold a small and manageable graph and synthesis this graph with trusted partner organisations.

VIVO [3] originated at Cornell University and was developed as an open source enterprise system with Funding by the US National Institutes of Health 2009-2012.   VIVO uses a collection of ontologies,  including  the  VIVO Integrated Semantic Framework (VIVO-ISF [4]) to  represent scholarship.   VIVO is in use at more than 150 sites in 26 countries.   VIVO data regarding scholarship can be queried using the W3C standard SPARQL query language, and has been pooled into a global search capability, CTSAsearch, hosted by the University of Iowa.

In this presentation, we demonstrate how the Research Graph data can be linked to 150 VIVO sites. In addition, we discuss, a new collaborative project between VIVO and Research Graph to have a rapid solution for launching VIVO instances from Research Graph data.

INTEROPERABILITY

In Open Repository Conference 2017 [2], the authors presented how to create an Open Linked Data model for Research Graph using VIVO ontology. We have extended this work by providing an automated pipeline for ingesting XML from Open  Access Repositories,  transform these records into   VIVO RDF, and integrate them into an operational VIVO instance. In addition, we are working on a rapid deployment platform that enables automated ETL between research data repositories (and open access repositories such as DSpace) to Research Graph. The vision is to link disconnected scholarly  works  to  the  global  network  of  scholarly communication, and we hope this work makes the research information more connected, discoverable and reusable. Figure 2 shows the pipeline for transforming DSpace records to a graph database and also make them accessible via VIVO instance.

Figure 2: From DSpace to Research Graph and VIVO

Although this is a work in progress, we aim to show a live demo of the pipeline presented in Figure 2.

CONCLUSION

In this presentation, we will report on the progress of the work on VIVO and Research Graph interoperability. This presentation builds on the prior work presented in the recent Open Repository Conference where we presented the initial mapping between VIVO RDF and Research Graph metamodel. Given the aim of this work is to provide production level solution for research management systems, we will demonstrate the outcome of our work in action, and we discuss the technical challenges, and the roadmap for future developments toward a sustainable and accessible solution.

REFERENCES

1. Amir  Aryani,   Data  Description  Registry  Interoperability  WG:  Interlinking  Method  and  Specification  of

Cross-Platform Discovery, Research Data Alliance, doi:10.15497/RDA00003

2. Conlon, Michael, Aryani, Amir, Creating an open linked data model for Research Graph using VIVO Ontology

Conference, Open Repositories CONFERENCE, 2017, Brisbane, Australia

3. Börner, K., Conlon, M., Corson-Rikert, J., Ding, Y. (eds.) VIVO: A Semantic Approach to Scholarly Networking

and Discovery, Morgan-Claypool, 2012. p. 1-175.

4. OpenRIF: VIVO integrated semantic framework data standard github repository. Web site (accessed on Nov 20,

2016), https://github.com/openrif/vivo-isf-ontology

Biography

Dr Amir Aryani is the co-chair of the Data Description Registry Interoperability WG in Research Data Alliance and the project lead for the Research Data Switchboard. He is working in the capacity of a project manager for Australian National University (ANDS), and part of this role is to manage ANDS interoperability projects with international partners. He has completed his PhD in the field of software evolution at the school of computer science, RMIT university, and he has peer-reviewed publications in fields of Software Engineering, Software Evolution and eResearch.

Dr Adrian Burton¹, Dr Amir Aryani¹, Ms Natasha Simons²,  Ms Catherine Brady¹

¹Australian National Data Service, Canberra, Australia,

²Australian National Data Service, Brisbane, Australia

SUMMARY

Links between scholarly literature and data enable discovery of and access to related knowledge and underpinning observations. They facilitate reuse, reproducibility and transparency of research. In practice, however, these links are difficult to find or share. The main reason for this is that there is no universal way of exchanging link information between databases and systems which hold this information. The international Scholix initiative aims to address this problem by providing an overarching framework for existing technical initiatives that individually address parts of the overall problem. Research institutions can join the Scholix effort by becoming contributors and consumers. Australian researchers contributing to the Scholix initiative via Research Data Australia gain increased exposure, and potentially recognition, for their data and linked publications through third party services like Scopus.

INTRODUCTION

Links between a journal article and data enable the reader of the article to follow the connection  to the data that supports the findings of the research. Conversely, they enable a user of a dataset to find literature based on that dataset. These links significantly aid the scientific method by improving discovery of research outcomes and access to related knowledge and underpinning observations. While there are clear benefits to literature and data linking, in practice these links are difficult to find or share. The main reason for this problem is the lack of a universal standard for exchanging connections between databases and systems which hold this information. Instead, there are different agreements and technical frameworks for exchanging links between different partners and scholarly systems managed by bilateral agreements between individual institutions. . The Scholix initiative[1] aims to address this problem. Its goal is to improve the connectivity between scholarly literature and research data as well as between datasets, thereby making it easier to discover, interpret and reuse scholarly information.

WHAT IS SCHOLIX?

Scholix is short for Scholarly Link Exchange. The goal of Scholix is to improve the links between scholarly literature and research data as well as between data and data. Scholix is global in scope and is an initiative of the Research Data Alliance (RDA) and the World Data System (WDS). It is supported by a number of partner organisations, including ANDS, with many people involved in its development.

The Scholix initiative offers:

1. a universal, global framework that enables information about the links between scholarly articles and data to be exchanged
2. technical guidelines that specify how the interoperability framework will work
3. a common conceptual model, an information model and open exchange protocols.

HOW DOES SCHOLIX WORK?

Scholix provides an overarching framework for existing technical initiatives that individually address parts of the overall problem that is hindering better linking between data and literature [2]. It also provides a conceptual model and an information model. Within the Scholix framework repositories, data centres, journals and others provide information about the links between literature and data that they hold to community ‘Hubs’ such as OpenAire, Crossref and DataCite. The community ‘Hubs’ – which are natural places to collect and exchange information about the links between literature and data – commit to a common information model for exchanging the links that they hold and an agreed open exchange method enables this to occur.

HOW CAN RESEARCH INSTITUTIONS CONTRIBUTE AND BENEFIT?

Research institutions can join the Scholix effort to increase the links between scholarly literature and data. The Research Data Australia Registry enables contributor metadata records containing connections between data collections and related publications to be mapped to the Scholix Framework’s Information Model [3]. The benefits of becoming a contributor include improving the discovery of the linked research objects and potentially increasing web traffic to the institution. The benefits of becoming a consumer include better tracking of datasets and publications plus making it easier for researchers and others to find and access relevant datasets. Australian researchers contributing to the Scholix initiative via Research Data Australia gain increased exposure, and potentially recognition, for their data and linked publications through third party services like Scopus [4].

PRESENTATION OUTLINE

This presentation will:

• introduce the international Scholix initiative to the Australasian eResearch community
• enable a discussion on how to link data and publications to enhance research discovery and reproducibility
• outline the reasons behind the Scholix initiative and its potential benefits to researchers
• provide detail on how Scholix works at a technical level
• demonstrate how research institutions can contribute to the Scholix initiative and gain benefits from participation

REFERENCES

1. Scholix: a framework for scholarly link exchange. Available from http://www.scholix.org/. Accessed 27 June 2017.
2. Burton, A. et al. The Scholix Framework for Interoperability in Data-Literature Information Exchange. D-Lib (Jan 2017). https://doi.org/10.1045/january2017-burton
3. Linking data with Scholix. Available from http://www.ands.org.au/working-with-data/publishing-and-reusing-data/linking-data-with-scholix. Accessed 27 June 2017.
4. New on Scopus: Link to datasets, search funding acknowledgements and find more CiteScore transparency. Available from  https://blog.scopus.com/posts/new-on-scopus-link-to-datasets-search-funding-acknowledgements-and-find-more-citescore Accessed 27 june 2017.

Biographies

Dr Adrian Burton is Director of Services at the Australian National Data Service (ANDS).  In this capacity he has a keen interest in national services that enable data publication, data discovery and data citation as well as the human support services that build the capability of researchers and research organisations to take advantage of data infrastructure. Adrian has provided strategic input into several national infrastructure initiatives, including Towards an Australian Research Data Commons, The National eResearch Architecture Taskforce, and the Australian Research Data Infrastructure Committee.   Adrian is active in building national policy frameworks to unlock the value in the research data outputs of publicly funded research.

Dr Amir Aryani is the co-chair of the Data Description Registry Interoperability WG in Research Data Alliance and the project lead for the Research Data Switchboard. He is working in the capacity of a project manager for Australian National University (ANDS), and part of this role is to manage ANDS interoperability projects with international partners. He has completed his PhD in the field of software evolution at the school of computer science, RMIT university, and he has peer-reviewed publications in fields of Software Engineering, Software Evolution and eResearch.