We have a long-standing expertise in developing data acquisition, semantic integration, and reuse frameworks, particularly for heterogeneous, multi-scale data sources. Relevant strengths include:

• Semantic interoperability. Extensive experience designing and implementing metadata to support cross-domain and cross-country data integration. “Semantic Modelling of Earth Observation / Remote Sensing” (https://doi.org/10.1016/j.eswa.2021.115838): semantic integration of satellite data.
• Data management and analysis. We have developed solutions to connect distributed databases and high-performance computing environments, enabling efficient data processing. “Scalable approach for high-resolution land cover: a case study in the Mediterranean Basin” (https://doi.org/10.1186/s40537-023-00770-z): big-data workflow processing > 4 TB of multi-source EO data covering 450 Sentinel-2 + 1200 ASTER tiles.
• Interoperability. Creation of reference datasets and data harmonization across institutions, aligned with FAIR principles. Creating harmonised datasets in collaboration with public administrations in the agri-food and environmental sectors, with the development of the Andalusian Data Space for the AgriFood Sector (https://edaan.agora-datalab.eu/en).
• Metadata standards, automated quality assessment & NLP tools. We develop metadata schemas, ontologies, and (semi-)automated data quality validation tools. We have developed metadata schemas and ontologies supporting automated quality checks. For instance, the ontology-driven KPI meta-modelling work demonstrates advanced metadata modelling (https://doi.org/10.1016/j.ijinfomgt.2019.10.003).
• Multi-layer geospatial analytics. Expertise in building geospatial data tools compatible with REST, GraphQL, and SPARQL APIs, integrating satellite data and sensor data for downstream analytics. With the land-cover mapping work, we integrate geospatial multi-source data, satellite imagery, and high-performance computing, offering APIs and geospatial services (https://ercim-news.ercim.eu/en140/special/knowledge-driven-strategy-for-scalable-land-cover-mapping-using-earth-observation-data).
• Service discoverability and composability. Our platform TITAN (https://doi.org/10.1016/j.knosys.2021.107489) provides a powerful environment for orchestrating data-based services, enabling reuse and dynamic composition by end-users and third-party companies.
• Satellite data exploitation for agriculture. From land-cover work to precision agriculture use cases, we utilize NDVI/NDWI indices and other derived features to use cases.
• Strengthening adoption of AI in agriculture. We work on explainable, trustworthy, and user-centric AI solutions, addressing barriers to adoption such as transparency, reliability, and decision traceability. For example, we integrated evolutionary ML and interpretability in real-world use cases (https://doi.org/10.1016/j.engappai.2024.109628).

Khaos Research is a research group at the University of Malaga specializing in data technologies, semantic interoperability, AI, and large-scale data management for agri-food and environmental domains.

We are very interested in joining a consortium that aims to address these challenges, contributing our expertise in:
• semantic data infrastructures
• FAIR data management
• geospatial analytics
• trustworthy AI for agriculture
• policy-related data engineering
• data-based service composition (via TITAN)

If this aligns with your ongoing plans or if you are forming a consortium, we would be pleased to discuss potential collaboration opportunities.

I would be happy to arrange a short meeting to explore synergies in more detail.