Data Discovery Platform with Generalized Exploratory, Management, and Search Capabilities

Información del proyecto

DataGEMS

Identificador del acuerdo de subvención: 101188416

DOI

10.3030/101188416

Fecha de la firma de la CE 25 Octubre 2024

Fecha de inicio 1 Enero 2025

Fecha de finalización 31 Diciembre 2027

Financiado con arreglo a

• Research infrastructures

Coste total

€ 6 688 960,00

Aportación de la UE

€ 6 688 960,00

Coordinado por

ATHINA-EREVNITIKO KENTRO KAINOTOMIAS STIS TECHNOLOGIES TIS PLIROFORIAS, TON EPIKOINONION KAI TIS GNOSIS
Greece

CORDIS proporciona enlaces a los documentos públicos y las publicaciones de los proyectos de los programas marco HORIZONTE.

Los enlaces a los documentos y las publicaciones de los proyectos del Séptimo Programa Marco, así como los enlaces a algunos tipos de resultados específicos, como conjuntos de datos y «software», se obtienen dinámicamente de OpenAIRE .

Resultado final

Other (5)

D4.3a User Assistance Services

This deliverable presents the first version of the User Assistance Services

D3.3a Data Discovery Services

This software deliverable will present the first version of the Dataset Discovery Services. It will provide implementation progress, blocking issues, and future plans for components dealing with (a) in-dataset discovery, (b) cross-dataset discovery, (c) extracting and utilising data patterns, and (d) packaging of dataset into meaningful bundles for the user.

D5.3a Dataset and Model Management Services

This deliverable presents the first prototype version of Dataset and Model Management Services. It will be based on state-of-the-art systems and will include basic functionalities for the MoMa and Polyglot Query engine for the execution of dataset discovery. It will provide support for at least 2 basic structured and semi-structured data-types, e.g., relational data and CSV files.

D6.3a Operational Services & Integration Components

This deliverable presents the first version Operational Services and Integration Components of the project platform

D2.3a Dataset Profiling Services

This deliverable presents the first version of the a) methods for generating quantitative and qualitative data insights, (b) adapters to generate semantic representations of a dataset, (c) methods for generating summaries in natural language and as graphs.In addition will present the first version of (a) multimodal methods to extract information from unstructured data (b) integrated with the semantic precision and grounding provided by KGs and online algorithms for processing streaming data, which may be structured or unstructured

Documents, reports (14)

Data Storage, Management and Access Architecture Report

This deliverable presents the design of the architecture for the Data Storage, management and Access components. It will describe the components to store datasets and their metadata, the machine learning models, the virtual knowledge graph layer, as well as the user profile, the polyglot engine, and performance monitoring components.

D1.4a Short policy brief on the expected impact from the project

The policy brief aims at assessing the impact of the implementation of DataGEMS across different sectors and providing stakeholders with actionable insights in relation to its use and adoption. The goal of this deliverable is checking the different platform capabilities against their benefits and challenges in multiple contexts including real-life use cases, in order to draft policy recommendations for optimising, improving and adapting DataGEMS, addressing any potential issues that may be identified such as privacy or developments on Artificial Intelligence, fostering further collaboration through public-private partnerships and ensuring that it can continue benefiting future generations.

User Assistance Architecture Report

This deliverable will present the whole architecture of the User Assistance algorithms and methods involving Query Recommendation, Query Disambiguation, Dataset Recommendation and Explanations

Dataset Profiling Design Report

This deliverable aims to provide the design of the Dataset Profiling for Structured Datasets, Unstructured Datasets as well as streaming data which may be structured or unstructured

Pilots Specifications & Design Report

This deliverable presents the Specifications and Design of the 4 Pilots

Dataset Discovery Architecture Report

This deliverable presents a full description of the dataset discovery architecture with well defined components responsible for allowing the user to exploit as efficiently as possible data provided by DataGEMs. The deliverable will describe how the project plans offering both in-dataset and cross-dataset discovery along with utilities such as exploiting data analytics patterns and packaging of datasets into bundles.

D7.4a Pilot's Software Components

This deliverable presents the first version of the Pilots' Software Components

Pilot Requirements Report

This deliverable presents the whole set of Requirements for the 4 Pilots

Awareness Raising, Communication, Dissemination, Outreach & Engagement Plan

This deliverable presents the Dissemination and Communication plan in terms of means and timeframe of activities, in combination with stakeholder identification and targeting. The Dissemination and Communication plan will be updated as necessary to ensure its relevance and effectiveness.

D8.3a Report on Dissemination, Communication and Liaison Activities

This deliverable depicts the process and the results of the establishment of liaison activities with relevant initiatives/taskforces/projects that operate in the areas of Open Science, EOSC and EU Research Infrastructures.

D1.5a Ethics Compliance

The goal of this deliverable is assessing ethical aspects and data implications of DataGEMS, including any data protection elements that may be relevant for the platform. This will be achieved through an ethical impact assessment that includes an evaluation of compliance with fundamental rights and main ethical principles, with a specific separate analysis for the use of Artificial Intelligence as applicable, taking into account the Ethics Guidelines for Trustworthy AI published by the AI High Level Expert Group and any provisions that may need to be observed according to the AI Act. A data compliance analysis will also be carried out, ensuring that DataGEMS complies with the applicable legal framework in this field, covering the GDPR, the Regulation on a framework for the free-flow of non-personal data in the EU and any concerned national laws.

Platform Architecture Design Report

This deliverable will present the Architecture Design of the Project Platform and its key specifications. It will describe the various components required, including their communication pathways and dependencies. Emphasis will be placed on providing a clear, diagram-based design of these components while ensuring flexibility for future modifications and additions.

Platform Development Manual

This deliverable will present the Development Manual of the Project Platform

D7.5a Pilot Implementation Evaluation Report

This deliverable presents the first version of the evaluation of the pilots implementation

Websites, patent fillings, videos etc. (1)

D8.2a Material for Awareness, Communication, Dissemination and Outreach

This deliverable deals with the delivery of visual and textual representation of the project to be used in the project’s dissemination activities, be it web, face-to-face or printed ones. It also establishes guidelines for promotion and use of materials delivered. Apart from templates, the task will deliver early in the project lifetime generic material for reuse in several face-to-face, printed and online occasions and in several languages.

Data Management Plan (1)

D1.3a Data Management Plan

This deliverable lays a solid roadmap for management of data, software, workflows, and any other research output expected by the project. Due to the data-centric nature of the project, the deliverable will describe in detail the whole lifecycle of data in DataGEMs including acquisition, storing, updating, and archiving along with a clear plan of ensuring accessibility and reusability of all project outputs.

Publicaciones

Peer reviewed articles (12)

Comparison and Analysis of Value Linking in Text-to-SQL systems: [Experiments & Analysis]

Autores: Apostolos Koukouvinis, George Katsogiannis-Meimarakis, Georgia Koutrika
Publicado en: Proceedings of the ACM on Management of Data, Edición 4, 2026, ISSN 2836-6573
Editor: Association for Computing Machinery (ACM)
DOI: 10.1145/3802029

$k$-Graph: A Graph Embedding for Interpretable Time Series Clustering

Autores: Boniol, Paul; Tiano, Donato; Bonifati, Angela; Palpanas, Themis
Publicado en: IEEE Transactions on Knowledge and Data Engineering, 2025, ISSN 1558-2191
Editor: IEEE
DOI: 10.48550/ARXIV.2502.13049

Graph-Based Vector Search: An Experimental Evaluation of the State-of-the-Art

Autores: Ilias Azizi, Karima Echihabi, Themis Palpanas
Publicado en: Proceedings of the ACM on Management of Data, Edición 3, 2026, ISSN 2836-6573
Editor: Association for Computing Machinery (ACM)
DOI: 10.1145/3709693

LeaFi: Data Series Indexes on Steroids with Learned Filters

Autores: Qitong Wang, Ioana Ileana, Themis Palpanas
Publicado en: Proceedings of the ACM on Management of Data, Edición 3, 2026, ISSN 2836-6573
Editor: Association for Computing Machinery (ACM)
DOI: 10.1145/3709701

Graphint: Graph-Based Time Series Clustering Visualisation Tool

Autores: Boniol, Paul; Tiano, Donato; Bonifati, Angela; Palpanas, Themis
Publicado en: 2025 IEEE 41st International Conference on Data Engineering (ICDE), 2025, ISSN 2375-026X
Editor: IEEE
DOI: 10.48550/ARXIV.2503.07698

DARTH: Declarative Recall Through Early Termination for Approximate Nearest Neighbor Search

Autores: Manos Chatzakis, Yannis Papakonstantinou, Themis Palpanas
Publicado en: Proceedings of the ACM on Management of Data, Edición 3, 2026, ISSN 2836-6573
Editor: Association for Computing Machinery (ACM)
DOI: 10.1145/3749160

APEROL: Adaptive Parallel Edge-to-Cloud Runtime Optimization for Layered Workflow Execution

Autores: Dimitrios Banelas, Alkis Simitsis, Nikos Giatrakos
Publicado en: Proceedings of the VLDB Endowment, Edición 19, 2026, ISSN 2150-8097
Editor: Association for Computing Machinery (ACM)
DOI: 10.14778/3773749.3773755

Model reusability in Reinforcement Learning

Autores: Nikookar, Sepideh; Namazi Nia, Sohrab; Basu Roy, Senjuti; Amer-Yahia, Sihem; Omidvar-Tehrani, Behrooz
Publicado en: The VLDB Journal, 2025, ISSN 0949-877X
Editor: Springer Nature
DOI: 10.1007/S00778-025-00920-0

Multi-objective Clustering Algorithm Applied to the MathE Categorization Problem

Autores: Beatriz Flamia Azevedo, Gabriel A. Leite, M. Fátima Pacheco, Florbela P. Fernandes, Ana Maria A. C. Rocha, Ana I. Pereira
Publicado en: Information Systems Frontiers, 2026, ISSN 1387-3326
Editor: Springer Science and Business Media LLC
DOI: 10.1007/S10796-025-10674-3

Subspace Collision: An Efficient and Accurate Framework for High-dimensional Approximate Nearest Neighbor Search

Autores: Jiuqi Wei; Xiaodong Lee; Zhenyu Liao 0001; Themis Palpanas; Botao Peng
Publicado en: Proceedings of the ACM on Management of Data, 2025, ISSN 1557-7341
Editor: PACMMOD
DOI: 10.48550/ARXIV.2411.14754

VUS: effective and efficient accuracy measures for time-series anomaly detection

Autores: Paul Boniol; Ashwin K. Krishna; Marine Bruel; Qinghua Liu; Mingyi Huang; Themis Palpanas; Ruey S. Tsay; Aaron J. Elmore; Michael J. Franklin; John Paparrizos
Publicado en: The VLDB Journal, 2025, ISSN 0949-877X
Editor: Springer Nature
DOI: 10.48550/ARXIV.2502.13318

R-Fairness: Assessing Fairness of Ranking in Subjective Data

Autores: Balzotti, Lorenzo; Firmani, Donatella; Mathew, Jerin George; Torlone, Riccardo; Amer-Yahia, Sihem
Publicado en: Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2025, ISSN 0736-587X
Editor: Association for Computational Linguistics
DOI: 10.5281/ZENODO.15504617

Conference proceedings (23)

Reducing Human Effort in Evaluating Small and Medium Language Models as Students and as Teachers

Autores: Oleh Prostakov, Viacheslav Hodlevskyi, Nassim Bouarour, Adam Sanchez-Ayte, Noha Ibrahim, and Sihem Amer-Yahia
Publicado en: 2025, ISSN 2150-8097
Editor: PVLDB

Optimizing Source Selection for Tuple-Value Discovery

Autores: Ahmad Fares, Georgia Troullinou, Silviu Maniu, Sihem Amer-Yahia
Publicado en: 2025, ISSN 2150-8097
Editor: VLDB

DeviceScope: An Interactive App to Detect and Localize Appliance Patterns in Electricity Consumption Time Series

Autores: Adrien Petralia, Paul Boniol, Philippe Charpentier, Themis Palpanas
Publicado en: 2025 IEEE 41st International Conference on Data Engineering (ICDE), 2025
Editor: IEEE
DOI: 10.1109/ICDE65448.2025.00350

Few Labels are All you Need: A Weakly Supervised Framework for Appliance Localization in Smart-Meter Series

Autores: Adrien Petralia, Paul Boniol, Philippe Charpentier, Themis Palpanas
Publicado en: 2025 IEEE 41st International Conference on Data Engineering (ICDE), 2025
Editor: IEEE
DOI: 10.1109/ICDE65448.2025.00329

An End-to-End Re-Evaluation of Table Entity-Linkers

Autores: Christensen Martin Pekár, Lissandrini Matteo, Hose Katja
Publicado en: 2026
Editor: Zenodo

User-Friendly Foundation Model Adapters for Multivariate Time Series Classification

Autores: Romain Ilbert, Vasilii Feofanov, Malik Tiomoko, Ievgen Redko, Themis Palpanas
Publicado en: 2025 IEEE 41st International Conference on Data Engineering Workshops (ICDEW), 2025
Editor: IEEE
DOI: 10.1109/ICDEW67478.2025.00024

Optimizing UDF Queries in SQL Data Engines

Autores: Konstsntinos Chasialis, Yannis Foufoulas, Alkis Simitsis, Yannis Ioannidis
Publicado en: 2026, ISSN 2367-2005
Editor: OpenProceedings

PG-HIVE: Hybrid Incremental Schema Discovery for Property Graphs

Autores: Sofia Sideri, Georgia Troullinou, Elisjana Ymeralli, Vasilis Efthymiou, Dimitris Plexousakis, Haridimos Kondylakis
Publicado en: 2026, ISSN 2367-2005
Editor: OpenProceedings

NILMFormer: Non-Intrusive Load Monitoring that Accounts for Non-Stationarity

Autores: Adrien Petralia, Philippe Charpentier, Youssef Kadhi, Themis Palpanas
Publicado en: Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2, 2026
Editor: ACM
DOI: 10.1145/3711896.3737251

Towards a Neural Database Execution Engine

Autores: Christos Tsapelas
Publicado en: 2025
Editor: CEUR-WS

Hyppo: Efficient Discovery and Execution of Data Science Pipelines in Collaborative Environments

Autores: Antonios Kontaxakis, Dimitris Sacharidis, Alkis Simitsis, Alberto Abelló, Sergi Nadal
Publicado en: 2025, ISSN 2367-2005
Editor: OpenProceedings

In-depth Analysis of LLM-based Schema Linking

Autores: George Katsogiannis-Meimarakis, Katsiaryna Mirylenka, Paolo Scotton, Francesco Fusco, Abdel Labbi
Publicado en: 2026, ISSN 2367-2005
Editor: OpenProceedings

Analysis of Text-to-SQL Benchmarks: Limitations, Challenges

Autores: Anna Mitsopoulou,Georgia Koutrika
Publicado en: 2025
Editor: OpenProceedings

Flood Detection in Switzerland Using Sentinel-2 Satellite Imagery

Autores: Claude Lehmann, Kurt Stockinger, Pavel Sulimov
Publicado en: 2026 IEEE Swiss Conference on Data Science and AI (SDS), 2026
Editor: IEEE
DOI: 10.1109/SDS70563.2026.00033

Multi-Modal Data Exploration via Language Agents

Autores: Farhad Nooralahzadeh, Yi Zhang, Jonathan Fürst, Kurt Stockinger
Publicado en: Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics, 2026
Editor: The Asian Federation of Natural Language Processing and The Association for Computational Linguistics
DOI: 10.18653/V1/2025.FINDINGS-IJCNLP.47

UDFB <scp>ench</scp> : A Tool for Benchmarking UDF Queries on SQL Engines

Autores: Yannis Foufoulas, Theoni Palaiologou, Alkis Simitsis
Publicado en: Companion of the 2025 International Conference on Management of Data, 2025
Editor: ACM
DOI: 10.1145/3722212.3725139

What Drives Learned Optimizer Performance? A Systematic

Autores: Kostas Mparmparousis, Christos Tsapelas, Georgia Koutrika
Publicado en: OpenProceedings, 2026, ISSN 2367-2005
Editor: OpenProceedings

The UDFBench Benchmark for General-purpose UDF Queries

Autores: Yannis Foufoulas, Theoni Palaiologou, and Alkis Simitsis
Publicado en: 2025
Editor: PVLDB
DOI: 10.14778/3746405.3746409

Problem Solved? Information Extraction Design Space for Layout-Rich Documents using LLMs

Autores: Gaye Colakoglu, Gürkan Solmaz, Jonathan Fürst
Publicado en: Findings of the Association for Computational Linguistics: EMNLP 2025, 2025
Editor: Association for Computational Linguistics
DOI: 10.18653/V1/2025.FINDINGS-EMNLP.973

HAIDES: Adaptive Approximation of Inference Queries over Unstructured Data

Autores: Christos C. Papadopoulos, Alkis Simitsis, Torben Bach Pedersen
Publicado en: 2025 IEEE 41st International Conference on Data Engineering (ICDE), 2025
Editor: IEEE
DOI: 10.1109/ICDE65448.2025.00181

Linear Transformations for Privacy-Preserving Machine Learning

Autores: Pavel Sulimov, Sergey Eremeykin
Publicado en: 2026 IEEE Swiss Conference on Data Science and AI (SDS), 2026, ISSN 2835-3420
Editor: IEEE
DOI: 10.1109/SDS70563.2026.00034

Supervised Cardinality Estimation for Similarity Queries on Knowledge Graphs

Autores: Georgios Chatzigeorgakidis, Dimitrios Skoutas, Alkis Simitsis
Publicado en: 2026, ISSN 2472-9671
Editor: IEEE

Evaluating and Generating Query Workloads for High Dimensional Vector Similarity Search

Autores: Matteo Ceccarello, Alexandra Levchenko, Ioana Ileana, Themis Palpanas
Publicado en: Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2, 2026
Editor: ACM
DOI: 10.1145/3711896.3737383

Other (1)

Jazero: A Semantic Table Search System

Autores: Christensen, Martin Pekár; Lissandrini, Matteo; Hose, Katja
Publicado en: 2026
DOI: 10.5281/ZENODO.18935699