19-23 September 2022
Prague, Czech Republic
Europe/Amsterdam timezone

Next generation of the EOSC Portal - ML/AI enhanced user interface

22 Sep 2022, 16:35
Opal, Topaz

Opal, Topaz


Krzysztof Martyn


EOSC is a pan-European initiative that offers access to resources and services that foster scientific research. It involves various stakeholders, from researchers, providers, facilitators, up to commercial users. As a result, EOSC provides variety of resources for an open science market in Europe. Currently, it provides access to diverse and large data sources, comprising research papers, access to specialized infrastructure, datasets, research projects etc. To find relevant and interesting items, the user must understand what they need quite well, either by specifying expressive search queries or navigating through ontologies of resources and services, which could be overwhelming. An AI/ML-enhanced recommender system will provide assistance by combining different sources of data together, offering each user a personalized and customized view on the resources that they could be interested in. These features would not only facilitate discoverability of resources offered in EOSC Portal, but also extract and exploit relevant relationships among them, to deliver concise, data-supported information to the user. It is particularly important both for large research institutes, and users representing a long-tail of science, i.e., not affiliated to large and well-funded organizations, conducting research niche domains, or just not aware of the existing European resources. Moreover, EOSC is a dynamic and evolving environment, any implemented feature has to take into account the possibilities of further co-creation and advances to address needs unforeseen at the inception. The demonstration will feature the overall user interface for EOSC, featuring a weighted search, based on various criteria. We will introduce the use-cases, concepts, and classifications of information in the next-generation user interface, we'll also demonstrate and explain the components and interactions of the RS, specifically with the outlook into flexibility of the overall system and possible evolution.

Topic Machine Learning/Artificial Intelligence

Primary authors

Krzysztof Martyn Raimundas Tuminauskas Mr Marcin Wolski (PSNC) Mr Bartosz Walter (PSNC)

Presentation Materials