CLOSE

Natural Language Processing

Publications, patent descriptions, R&I project objectives and deliverables, policy documents, public procurement, etc. contain a wealth of textual information detailing current challenges, proposed or demonstrated progress and the expected impact of the innovation process. The latest developments in the field of Natural Language Processing (NLP) offer new means for exploiting this semantic richness

and to characterise research portfolios and ii. societal needs and challenges to support strategic decision making.

NLP approaches are powerful tools for mapping scientific and technological domains because they allow to analyse individually each document related to science and technology, avoiding potential confusion/restrictions related to fixed taxonomies.

Our work with SDGs

See how the NLP technologies developed by SIRIS have allowed us to map how the European commission is funding research beneficial for the 17 UN's Sustainable Development Goals!

Close

Virtual Knowledge Graph data access and integratrion

Research and innovation data is scattered across different data sources, which are typically not interoperable with one another. Today, UNiCS is an Open Data platform based on Semantic technologies that integrates Open Data repositories about Higher Education, Research and Innovation in Europe.

These integrated data can be openly accessed through a SPARQL endpoint. UNiCS offers analytical tools and strategic insights to explore and contextualise data about Higher Education, Research and Innovation.

Our research has been funded by the European Open Data Incubator to develop UNiCS, an open platform based on semantics technologies, which integrates heterogenous data on research, development and innovation!

Close

Semantic Characterisation of Research Portfolios

Science, Technology and Innovation activities are today characterised by research trends in emerging domains that appear and blossom at ever faster rates. To characterise those trends, it is necessary to quickly and efficiently build semantic maps that define the breadth and depth of these emerging research niches. At SIRIS, we are developing methods based on the

combination of deep learning techniques and the exploitation of domain ontologies to rapidly build ad-hoc controlled vocabularies that define the semantic boundaries of given research areas of interest. In parallel, we develop techniques to preprocess textual records and tag them with respect to the terms featured in the vocabularies we craft.

The tools we produce to map Research Portfolios are openly available on the Zenodo Platform!

Close

Advanced Data Visualisation

Science, Technology and Innovation activities are today characterised by research trends in emerging domains that appear and blossom at ever faster rates. To characterise those trends, it is necessary to quickly and efficiently build semantic maps that define the breadth and depth of these emerging research niches. At SIRIS, we are developing methods based on the

combination of deep learning techniques and the exploitation of domain ontologies to rapidly build ad-hoc controlled vocabularies that define the semantic boundaries of given research areas of interest. In parallel, we develop techniques to preprocess textual records and tag them with respect to the terms featured in the vocabularies we craft.

SIRIS developed RIS3-MCAT for the Catalan government, an interactive platform that allows one to monitor trends and collaborations within the Catalan research ecosystem.

Research and Development

We recognise the importance of constantly questioning the world, shifting paradigms and reinventing oneself. For this reason, a dedicated space for R&D is safeguarded and cultivated at the core of the company. Text mining, data visualisation, and semantic technologies, are some of the key R&D topics for an enduring and sustainable success.

Research carried out at SIRIS aims at offering tools and methods to ease informed decision making in the Science, Technology and Innovation sector. As such, we perform research in several areas linked with data integration, exploitation, analysis and visualisation.

Dictionary definitions

Natural
Language
Processing

We develop AI models and technologies that can help make sense of and to navigate large portfolios of Science Technology and Innovation documents

Read More
Crossroads

Virtual Knowledge Graph data access and integration

We perform research on how to design and apply domain knowledge to ease data access by means of semantic web technologies

Read More
Molecules

Semantic Characterisation of Research Portfolios

We study technologies to build interactive and compelling maps of the Research Output of entire Research Ecosystems

Read More
data visualisation graph


Advanced Data Visualisation

We aim at producing aesthetically pleasant yet useful and tailor-made data visualisations, to ease an informed decision-making process

Read More

Funding

To fund our research activities, we have won several competitive grants awarded by national and European funding bodies.
SciLake (2023-2025)
Industrial PhD for Mapping Thematic Innovation in Research Portfolios (2022-2025)
Industrial PhD on NLP for the classification of science, technology and innovation literature (2022-2025)
INODE (2020-2023)
Mapping of research portfolios and their alignment with societal challenges (2021-2022)
EPNet (2014-2019)
OKHAERS (2017-2018)
Analysis and characterisation of patterns of interaction in knowledge production (2016-2018)
Development of a Semantic Integration System for multi-format data sources (2016-2018
UNICS (2016-2018)