2023-10-05, 11:50–11:55, Poster presentation
We demonstrate a novel web service that offers comprehensive spatial context for areas of interest. The spatial context includes diverse types of information, e.g., land coverage, demographic and econometric data, weather, and air quality measurements, and more. These datasets originated from numerous official sources and services across the EU, assuring the reliability and accuracy of the information. The spatial-context (SC) service is enhanced with a map-centric UI that serves as an input and output interface. The user can simply draw an arbitrary polygon within EU boundaries to gather aggregated information, visualize data, and select variables to render clusters and outliers. The integration of various data sources and the provision of a user-friendly interface makes SC service a powerful tool for analyzing and interpreting spatial data in a comprehensive and efficient manner.
The primary aim of this service is to enrich with additional metadata the GEOSS Portal cognitive search developed in the context of the EIFFEL EU-funded project. In particular, a specific spatial dataset can give a broader perspective in the context of additional multidiscipline data. For example, the physical and social surroundings can potentially influence the interpretation of the data and can provide insights into relationships among variables and underlying trends. Realizing the potential and perspective of this venture, we propose the SC service as an autonomous open-source application, able to provide insights quickly and straightforwardly using aggregated open datasets easily obtained from reliable sources, such as the Eurostat and Copernicus.
One of the significant characteristics of the proposed SC service is the ease with which their data sources can be extended. Registered users can upload almost any spatial dataset, specifying the necessary information for data integration into the service and making the data publicly available. Beyond providing aggregated results and data visualization, the service has been enriched with some extra features. The first is Machine Learning (ML)-based geospatial clustering, with clusters formed based on the user-selected variables. Users also have the option to upload ephemeral datasets to be combined with existing ingested sources. Clustering procedure results in an interactive choropleth map with cluster visualization, including the automatic detection of spatial outliers. Another available analysis tool will be the geographically weighted regression (GWR) operating multiple local regressions on point data. Furthermore, a spatiotemporal analysis can be performed for variables with data available for multiple years. This is achieved after estimating and clustering the GWR coefficients separately for each year. The modular architecture of the SP service allows for more features to be included in the future.
The work described in this abstract is part of the EIFFEL European project. The EIFFEL project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 101003518. We thank all partners for their valuable contributions.
As an EU R&D and Innovation Project Manager in LIBRA AI Technologies, Georgios manages the company's tasks in EU-funded projects.
He acquired his diploma at the School of Chemical Engineering of the National Technical University of Athens (NTUA). He holds a PhD in Economics of Innovation and Technology from the same university. During his studies, he deepened his technical knowledge by participating in summer programs at the University of Foggia in Italy and the University of Essex in the UK.
As a researcher, he has many publications in scientific journals and international academic conferences. He is a co-author of two scholarly books on entrepreneurship and project management. His management experience is enriched by participating in various EU and national research projects.
In LIBRA AI Technologies, Georgios works closely with the project management teams, the relevant stakeholders, and the company's clients to understand their needs and ensure that each project meets its initial expectations.
As a data scientist, Eleanna analyses and visualizes customers’ data and implements machine learning techniques for time-series, computer vision, NLP and geospatial data.
With a MEng in Electrical and Computer Engineering and MSc in Data Science and Knowledge Technologies, she is eager to learn and practice her theoretical knowledge to solve real-life problems in various areas from visualization techniques and data engineering task to deep learning.
Dr Yannis Kopsinis is a Libra AI Technologies co-founder and CEO. He received his PhD from the Dept. of Informatics and Telecommunications, Univ. of Athens, in 2004. He has worked on Machine learning theory, applications, consultancy and hands-on tasks for more than two decades in applications ranging from telecoms to audio and medical to digital marketing. He has gained several prestigious research grants, such as a Marie Curie IEF and a Ramón y Cajal Fellowship, the University of Granada, Spain. He has also worked for over five years as a senior research fellow in the School of Engineering and Electronics at the University of Edinburgh, UK. He has published over 70 papers in technical journals and conferences and co-authored three book chapters. His work has received more than 1500 citations.