EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An Ontology-driven Hebbian Associative Memory Framework for Semantic Enrichment and Interoperability of Earth Observation Data

Aliya Fahmi

Journal of Intelligent Decision Making and Information Science · 09.09.2026 · 3 (18s) · 1986-2014

DOI: 10.17762/jidmis.v3.5906 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Intelligent Decision Making and Information Science, online erschienen am 09.09.2026. Verfasst von Aliya Fahmi. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Earth observation data are described by a variety of formats, metadata structure and have complex semantics that puts challenges on knowledge representation, semantic enrichment and interoperability with various earth observation and geospatial data sources. Current data integration techniques are mostly based on classic metadata matching and static knowledge representation, which are not suitable to represent dynamic relations between the different heterogeneous concepts and observations in the environment. This paper presents an Ontology-Driven Hebbian Associative Memory Framework which is based on the Hebbian principle of associative learning and aims at semantically enriching and interoperating earth observation data. The proposed framework combines the ontology-based knowledge representation system and the Hebbian associative memory system to find, learn and retain meaningful relationships between observations, metadata concepts, and entities in the environment. The first is to build an ontology-based semantic representation to structure multiple Earth observation information and to set relationships between the information attributes, geographical entity, temporal attributes and environmental phenomena. The Hebbian associative memory mechanism is used to learn co-occurrence and semantic associations between the concepts, so as to be able to efficiently retrieve and infer related knowledge. The learned associations are then used to enhance metadata, address semantic relationships and enable interoperability between the various Earth observation data types. The proposed framework presents an adaptive knowledge representation mechanism that can enhance the semantic consistency of knowledge, information discoverability and cross-source data integration. Experiments are conducted using heterogeneous Earth observation and satellite data that prove the success of the proposed method for semantic enrichment, association discovery and interoperability, compared to traditional metadata-based and non-associative methods. The framework offers a potential model for smart and knowledge-intensive management of Earth observation data, and can be used in the fields of Earth observation for environmental monitoring, climate analysis, disaster forecasting and planetary exploration. It presents the results of the ESA research project, Ontology-driven Knowledge Representation for Hebbian Associative Memory, which aims to enhance and enrich Earth Observation data by incorporating their associated metadata.

Zitieren

APA

Fahmi, A. (2026). An Ontology-driven Hebbian Associative Memory Framework for Semantic Enrichment and Interoperability of Earth Observation Data. Journal of Intelligent Decision Making and Information Science, 3(18s), 1986-2014. https://doi.org/10.17762/jidmis.v3.5906

Vancouver

Fahmi A. An Ontology-driven Hebbian Associative Memory Framework for Semantic Enrichment and Interoperability of Earth Observation Data. Journal of Intelligent Decision Making and Information Science. 2026;3(18s):1986-2014. doi:10.17762/jidmis.v3.5906

BibTeX
@article{Fahmi2026,
  author = {Fahmi, Aliya},
  title = {An Ontology-driven Hebbian Associative Memory Framework for Semantic Enrichment and Interoperability of Earth Observation Data},
  journal = {Journal of Intelligent Decision Making and Information Science},
  year = {2026},
  volume = {3},
  number = {18s},
  pages = {1986--2014},
  doi = {10.17762/jidmis.v3.5906}
}
Kostenlos starten →