Big data, online sharing platforms, collaboration tools. Technology provides many opportunities for researchers to work together and build on each other’s successes.
This statement was explored at various sessions during the EuroScience Open Forum (ESOF 2018) in July in Toulouse, France. Here are four innovative ideas that stuck with us long after the closing ceremony.
1. “Our world needs strong multidisciplinary research to solve problems.” – Liz Marchant, PhD, Taylor and Francis, during the session “Tech Trends 2022”
In theory, there is general agreement on the high value of the synergy between “hard sciences” and “other sciences.” In practice, inclusion efforts focus mainly on biomedicine and life sciences, leaving social sciences and the arts behind. Technology has the potential to change that, facilitating collaboration through new channels and tools for conducting cross-disciplinary experiments. Scholarly sharing networks such as Mendeley and SSRN already provide a mechanism for researchers from different disciplines to comment on and bolster each other’s work. Researchers who use these tools to develop an interdisciplinary mindset and approach may start to see greater success than those who choose not to reach outside their own discipline.
2. “We need AI as a scientific partner” – Paul Groth, PhD, Elsevier, during the session “Increasing scientific productivity through AI”
For researchers to do their best work, they need to be able to build on what’s gone before – but “what’s gone before” is increasing all the time. A study published in 2015 revealed that in order to keep up with the cancer literature alone, “one would have to read 17 articles per waking hour (with a generous 8 hours per day for sleep and other personal activities) —365 days per year.” This is where AI can make a difference for researchers. The more research is published on a particular topic, the faster the process of extracting and curating information should happen to maintain a high level of accuracy and productivity while avoiding duplication. Already machine learning technologies play a role in delivering personalized recommendations, but the future role for the organizations that manage scientific information will be about bringing the right piece of information to the researcher at the right point in their workflow.
3. “Engage with the public, and don’t fear the public.” – Síle Lane, PhD, Sense about Science, during the session “Responsible science communication”
In the era of “fake news” and the proliferation of predatory journals, high quality science is more relevant than ever. Research can inform the decisions of individuals, governments and organizations and help them steer society to a successful future.
When the language of journal articles is not accessible to the public, research results are in danger of being misinterpreted or even ignored. To avoid such issues, the dissemination of research results should extend beyond the article itself and be done in a responsible way. To quote Rolf van Wegberg of the Network of the Netherlands (PNN), “Scholarly communication doesn’t stop when you publish; it starts when you publish.”
In line with the principles of open science, funders and institutions should encourage authors to include communication plans in the research cycle to ensure that their findings are understood and as widely shared and impactful as possible.
4. “We need AI to manage the large number of research objects coming in so the quality remains high and the integrity of research is not sacrificed.” –Ijsbrand Jan Aalbersberg, PhD, Elsevier, during the session “Tech Trends 2022”
What is the best way to counter the peer review quantity problem generated by the increasing number of publications and research elements? The conclusion reached during the TechTrends 2022 session was that scientists should be given rewards and incentives for transparency. Research outputs should not be the only evaluation criteria for a researcher. Furthermore, AI has the potential to assist the peer-review process by determining the source of data and detecting “fabrication” of data and plagiarism. While it must not, and will not, replace humans at the core of the peer-review process, it should help create and track the right metrics for the evaluation process.