Paper deep dive
Improving Interoperability among Defence and National Security Ontologies: Analysis and Evaluation Tasks
Jonathon Dilworth, Pedro Giesteira Cotovio, David Herron, Paul Cripps, Nigel Dewdney, Catia Pesquita, Ernesto Jiménez-Ruiz
Intelligence
Status: succeeded | Model: Gemma-4-26B-A4B | Prompt: intel-v1 | Confidence: 89%
Last extracted: 8/8/2026, 4:15:06 AM
Summary
This paper introduces the Defence, Intelligence and Security Ontologies (DISO) network, a collection of over 60 publicly available ontologies in the defence and national security domain. The authors analyze interoperability challenges and introduce a new track for the Ontology Alignment Evaluation Initiative (OAEI). This track includes eight matching tasks, consensus alignments derived from state-of-the-art systems, and manually-curated silver-standard mappings to facilitate ontology integration.
Entities (10)
Relation Signals (7)
DISO â evaluatedby â OAEI
confidence 95% · In this work, we analyse and document over 60 publicly available ontologies and introduce a new track for the Ontology Alignment Evaluation Initiative (OAEI).
LogMap â usedfor â DISO
confidence 92% · We conducted a bulk ontology alignment exercise using the ontology alignment system LogMap over the DISO ontologies...
DISO â contains â IES
confidence 90% · DISO contains the IES version released in November 2025.
DISO â contains â D3FEND
confidence 88% · The DISO collection includes 15 such ontologies... A prominent example is the MITRE Corporation D3FEND cyber-security ontology.
DISO â contains â JC3IEDM
confidence 85% · The DISO collection contains two authoritative information exchange ontologies... The other information exchange ontology in DISO is JC3IEDM...
GUARD â fundedby â The Turing Defence & Security Grand Challenge
confidence 85% · GUARD has been funded by The Turing Defence & Security Grand Challenge...
DISO â contains â STIX
confidence 82% · Table 3 lists STIXcybersecurity as a source ontology in the DISO-OAEI track tasks.
Cypher Suggestions (0)
No Cypher suggestions yet.
Abstract
Abstract:The use of ontologies and knowledge graphs is becoming increasingly widespread in the defence and national security domain. Numerous ontologies have been developed through initiatives led by academia, industry, and government. Achieving interoperability across diverse defence and national security ontologies remains a major challenge due to the domain's breadth and specialisation. In this work, we analyse and document over 60 publicly available ontologies and introduce a new track for the Ontology Alignment Evaluation Initiative (OAEI). This track comprises eight matching tasks, consensus alignments and manually-curated (silver-standard) mappings. The consensus alignments are derived by aggregating the outputs of several state-of-the-art ontology alignment systems. The silver-standard is obtained from the manual validation of the consensus alignment together with a subset of the unique mappings (i.e., mappings suggested by only one system).
Tags
Links
- Source: https://arxiv.org/abs/2608.05867v1
- Canonical: https://arxiv.org/abs/2608.05867v1
Trouble viewing inline? Open PDF directly â
Full Text
62,958 characters extracted from source content.
Expand or collapse full text
Improving Interoperability among Defence and National Security Ontologies: Analysis and Evaluation Tasks Jonathon Dilworth â ,1[0009â0009â0260â0492] , Pedro Giesteira Cotovio â ,1,2[0000â0001â6724â899X] , David Herron â ,1[0009â0008â2736â6789] , Paul Cripps 3 , Nigel Dewdney 4 , Catia Pesquita 2[0000â0002â1847â9393] , and Ernesto JimĂ©nez-Ruiz â1[0000â0002â9083â4599] 1 City St Georgeâs, University of London, London, UK ernesto.jimenez-ruiz@citystgeorges.ac.uk 2 LASIGE, Faculdade de CiĂȘncias, Universidade de Lisboa, Portugal 3 Defence Science and Technology Laboratory, UK 4 The Alan Turing Institute, UK Abstract. The use of ontologies and knowledge graphs is becoming increas- ingly widespread in the defence and national security domain. Numerous on- tologies have been developed through initiatives led by academia, industry, and government. Achieving interoperability across diverse defence and national se- curity ontologies remains a major challenge due to the domainâs breadth and spe- cialisation. In this work, we analyse and document over 60 publicly available ontologies and introduce a new track for the Ontology Alignment Evaluation Ini- tiative (OAEI). This track comprises eight matching tasks, consensus alignments and manually-curated (silver-standard) mappings. The consensus alignments are derived by aggregating the outputs of several state-of-the-art ontology alignment systems. The silver-standard is obtained from the manual validation of the con- sensus alignment together with a subset of the unique mappings (i.e., mappings suggested by only one system). Resource Type: Dataset and benchmark. DISO Repository: https://github.com/city-artificial-intelligence/diso OAEI Track: https://city-artificial-intelligence.github.io/diso-oaei/ Codes: https://github.com/city-artificial-intelligence/DISO-mappings Permanent repository: https://doi.org/10.5281/zenodo.20059506 License: MIT License Keywords: Ontology Matching · Ontology Alignment · Network of Ontologies · Defence & National Security · Evaluation · OAEI 1 Motivation Defence and national security is one of the main UKRI 1 priorities, where AI tech- niques can play a crucial role in preventing and mitigating threats. Indeed, the work â These authors contributed equally to this work. â Corresponding author. 1 UK Research and Innovation (UKRI) funding body. arXiv:2608.05867v1 [cs.AI] 6 Aug 2026 2J. Dilworth et al. presented in this paper is conducted within the scope of the GUARD project, 2 which focuses on Ensuring Interoperable and Trustworthy Knowledge Graphs for Defence and National Security AI. GUARD has been funded by The Turing Defence & Security Grand Challenge, a collaborative effort towards meeting the strategic requirements of UK Government agencies for systematic sensemaking at scale and pace. The development of hybrid learning and reasoning systems in general, and Neu- rosymbolic (NeSy) AI systems [17] in particular, is gaining increasing attention to overcome the challenges of purely data-driven models (e.g., LLMs - Large Language Models [44]) with respect to fairness, privacy, correctness, data and energy efficiency, identifiability, and eXplainability (XAI). In a domain like defence and national secu- rity, the design and development of robust, reliable and semantically sound AI mod- els is paramount to support the notion of systematic sensemaking. Knowledge Graphs (KGs) [25] play a key role in orchestrating diverse and heterogeneous data sources, while providing a semantic and mathematical (i.e., logic-based) representation of the domain. KGs are becoming essential components of NeSy systems [23] to (i) increase the coverage, validity and quality of the available data, (i) impose constraints during the learning process, (i) reason about what has been learned (i.e., validity of the pre- dictions), and, ultimately, (iv) enable well-informed decision-making. One of the main challenges when working with KGs is the definition of an ontology to model the domain. The domain for a defence and national security application can be, however, quite broad as it may require the interplay of several subdomains to cover, among others, threats (e.g., physical and cyber), relevant actors (e.g., government, ter- rorist groups, intelligence services), operations (e.g., military, emergency response), infrastructure (e.g., airport, bases), policies (e.g., defence strategy, international law), and geolocations. There are several contributions from academia, industry, and government to cre- ate ontologies and knowledge graphs for security and defence. Prominent examples are the general-purpose models IES 3 and DoDAF. 4 IES (Information Exchange Stan- dard) has been developed by the UK Government; while the DoDAF Formal Ontology has been implemented by the Department of War/Defense Chief Information Officer and serves as a NATO/multi-national model for defence. The IES models embrace se- mantic web technologies to represent general domain concepts such as locations, legal events, states, and measures. MITRE, funded by the National Security Agency, has also launched a family of cybersecurity ontologies to formalise offensive and defensive tech- niques [30]. 5 The Department of Homeland Security has also designed a core Ontology for the maritime domain. 6 Efforts from academia include ontologies for military (e.g., [11,60,35,41]), national security (e.g., [8,42,7]), and cybersecurity (e.g., [46,47,59,57]). There are also prominent examples of collaborations among government, academia and industry. For example, the Intelligence Community (IC) Ontology Working Group (DIOWG), from the US Department of Defense, is leading on the development and 2 GUARD: https://ernestojimenezruiz.github.io/projects/guard/ 3 https://github.com/IES-Org/ont-ies 4 https://dodcio.defense.gov/Library/DoD-Architecture-Framework/ 5 https://d3fend.mitre.org/ 6 https://bit.ly/ontology-maritime-domain Improving Interoperability among Defence and National Security Ontologies3 implementation of a National Security Ontology Foundry (NSOF), which will rely on the Basic Formal Ontology (BFO), 7 and the Common Core Ontology (CCO) 8 as their baseline standards. 9 BFO is an academia-driven upper-level ontology, widely adopted in life sciences ontologies (e.g., OBO Foundry 10 ), but with an increasing presence in government-related ontologies. NATO are also actively investigating the application of ontologies in the Command and Control space. As part of their work on the Data Centric Reference Architecture for the Alliance (DCRA), 11 they have outlined the need for a domain upper ontology for defence and security (CXCSRM) and ontology management services, with objectives closely aligned to those of the NSOF. The interoperability among ontologies at the upper- and mid-level has been facili- tated by efforts like IES, BFO and CCO. However, although IES and BFO are, in prin- ciple, compatible, a complete and formal alignment still needs to be established [4]. Furthermore, achieving interoperability across application ontologies may be particu- larly challenging in domains that require reconciling both the wide-ranging scope of concepts and the highly specialised representations in the subfields. Defence and na- tional security is an example of such domains. An ontology foundry for defence and national security, as the one proposed by the DIOWG, 12 should ensure that its appli- cation ontologies achieve interoperability, comprehensive coverage, and high quality. Without these guarantees, the deployment of ontologies and KGs risks limiting their effectiveness in supporting downstream defence and national security tasks. In this paper, we focus on the analysis of the interoperability among defence and national security ontologies. Our contributions are summarised as follows. (i) We have collected and documented 60+ public ontologies relevant to the defence and national security domain. (i) We have analysed their intersection by performing ontology align- ment over 1,653 ontology pairs. (i) We have designed a new OAEI track including 8 matching tasks where we have compared the outcomes of several state-of-the-art align- ment systems, created a consensus-based alignment, and manually verified a silver- standard reference alignment. The silver-standard has been generated by manually cu- rating the consensus alignment and a subset of the unique system-generated mappings (i.e., those suggested by only one system). The rest of the paper is organised as fol- lows. Section 2 introduces the Defence, Intelligence and Security Ontologies (DISO) network, and the motivation for gathering and integrating these ontologies. The new DISO-OAEI evaluation track is described in Section 3, together with a comparison among the state-of-the-art systems. Section 4 summarises the created resources. Finally, future work directions and conclusions are given in Section 5. 7 https://basic-formal-ontology.org/ 8 https://github.com/CommonCoreOntology/CommonCoreOntologies 9 https://bit.ly/us-dod-ontology-news 10 https://obofoundry.org/ 11 NATO. Data Centric Reference Architecture for the Alliance (2025): https://nhqc3s. hq.nato.int/apps/DCRA_Report/ 12 Recently rebranded as NSOWG (National Security Ontology Working Group). 4J. Dilworth et al. Table 1: A view of the conceptual structure of the DISO collection of defence, intelli- gence and security ontologies. The clusters (or subdomains) of the DISO collection are listed together with the number of ontologies assigned to each. DISO Cluster/Subcluster NameCluster SizeSubcluster Size agentic1 ambient intelligence1 context awareness3 cyber-security15 digital twins1 information exchange2 information security5 mid-level12 risk management1 robotics2 situation awareness4 smart environments8 » smart buildings4 » smart cities2 » smart homes2 upper-level8 totals63 2 Defence, Intelligence and Security Ontologies (DISO) DISO (Defence, Intelligence and Security Ontologies) 13 is a collection of 60+ publicly available Web Ontology Language (OWL) [9] ontologies related to the defence and na- tional security domain. The ontologies in DISO were identified and obtained during an ontology search exercise undertaken primarily during Nov/Dec of 2025, drawing on the academic literature (including surveys of cyber-security ontologies, e.g., [36,52,53]), government and standards-body materials, and public ontology registries and repos- itories. An ontology was included in the collection if (i) it is publicly available, (i) it is provided in (or has an available serialisation in) OWL, and (i) it is relevant to the defence and national security domain, either directly (i.e., developed by or for de- fence, intelligence and security organisations) or indirectly, through subdomains, such as âsituation awarenessâ or âsmart environmentsâ, that underpin defence and national security applications (as discussed below). Relevant ontologies distributed without a permissive (or confirmed) licence are not redistributed, but linked from their original locations. The provenance of each ontology (upstream source and licence) is recorded in the DISO repository, and future extensions of the collection are expected to follow the same criteria through a documented contribution protocol. The DISO collection has been clustered into defence and national security cate- gories (or subdomains). Table 1 shows the 11 clusters (or subdomains) of the DISO collection in alphabetic order along with the number of ontologies assigned to each 13 https://github.com/city-artificial-intelligence/diso/ Improving Interoperability among Defence and National Security Ontologies5 Fig. 1: A view of the conceptual overlaps that exist between some of the clusters (or subdomains) of the ontologies within the DISO (Defence, Intelligence and Security Ontology) collection. cluster. Three factors make it difficult to provide precise counts of the number of on- tologies in the DISO collection and within its component clusters. First, many of the on- tologies obtained for DISO are physically represented as networks of multiple compo- nent ontology files. SOUPA [6], for example, an ontology for ubiquitous and pervasive computing whose development was partly funded by DARPA, and which is assigned to the DISO cluster âsituation awarenessâ, is presented as a network of 18 component ontology files. For DISO ontology counting purposes, however, we count SOUPA as a single ontology. Hence, in this respect, the ontology counts reported in Table 1 sig- nificantly understate the number of physically distinct ontology files present within the DISO repository. Second, to make it easy for researchers to reuse such networked DISO ontologies, we have frequently merged such networks into single, all-inclusive ontol- ogy files. Third, some ontologies within DISO fit naturally into more than one DISO cluster. For example, the SOUPA ontology is regarded as being a member of both the âsituation awarenessâ cluster and the âcontext awarenessâ cluster. Figure 1 illustrates the conceptual overlap that exists between the clusters (subdomains). Regarding the FAIRness [18] of DISO , we have created metadata in RDF, extending the VoID and Dublin Core Terms vocabularies, to enhance the exploration of the DISO clusters and ontologies. The metadata includes links to the ontologies (remote and local versions), as well as their license. We have also added the repository to the LoD cloud. 14 to improve its findability. The DISO clusters and ontologies are also documented in our GitHub repository and summarised as follows: Situation awareness. Situation awareness [12,39] has to do with perceiving and under- standing the elements of a dynamic environment, and the relations between them, whilst 14 https://lod-cloud.net/dataset/DISO 6J. Dilworth et al. considering both space and time, and the prediction of the environmentâs future state(s). The generality of the notion of situation awareness is reflected in the number of other DISO clusters with which it has close conceptual associations. Context awareness. âContext awarenessâ, for instance, has primarily to do with in- dividuals and their immediate surroundings: their nearby environmental context. The common thread (and motivation) shared by the ontologies in this cluster is the idea of pervasive computing, where the objective is for mobile application services to be able to adapt their behaviour dynamically based on an application userâs perceived current environmental context. Thus, the notion of context awareness can be regarded as a spe- cialised form of situation awareness, where the situation of interest is the surroundings of a mobile, software application user. Smart environments. It is similarly reasonable to regard smart environments (whether they be homes, or buildings, or cities) as being specialised forms of situations: built en- vironment situations of various size and scale. The notion of âsmartâ links to the notion of âawarenessâ because the ontologies of the DISO âsmart environmentsâ cluster gener- ally presume some amount of sensor-based monitoring of the environment in question. For example, the DISO âsmart homesâ ontology ThinkHome [51] has an emphasis on the monitoring and control of home energy use. Ambient intelligence. Ambient intelligence [3] models devices that seamlessly intercon- nect and collaborate with each other as well as with human users. Like context aware- ness, ambient intelligence supports the idea of pervasive computing, but with more attention to supporting ideas like the Internet of Things (IoT). BOnSAI [55] is a smart building ontology for ambient intelligence, and is assigned to both the âsmart buildingsâ and âambient intelligenceâ DISO clusters. Digital twins. Ontologies are well-suited to supporting the realisation of digital twins [38], and their use in connection with digital twins has been examined in a recent survey [31]. The notions of situation awareness, smart environments and ambient intelligence relate strongly to the notion of digital twins. Indeed, what DISO refers to as smart environ- ments (smart homes, buildings and cities) are prime examples of the types of physical things that the UKâs National Digital Twin Programme (NDTP) 15 envisages having digital twins. SAREF (a Smart Applications REFerence ontology) 16 , created by the Eu- ropean Telecommunications Standards Institute (ETSI), is one of the main ontologies discussed in connection with creating digital twins for buildings in [38]. Information security. Information security has to do with protecting information in all its forms, whether physical or digital, tangible (e.g., paperwork) or intangible (e.g., knowledge), from things like unauthorised access or inappropriate use, modification, disclosure or corruption. DISO contains five ontologies covering this domain. One of these, MDISOnt [40], a multi-dimensional information security ontology, decomposes information security into several perspectives using dimensional views and modules. Cyber-security. Cybersecurity is widely understood as a specialised form of information security that focuses on the digital domain and the protection of assets against digital threats. Three recent surveys of cyber-security ontologies [36,52,53] attest to the fact 15 https://ndtp.co.uk 16 https://saref.etsi.org/ Improving Interoperability among Defence and National Security Ontologies7 that the cyber-security domain appears to appeal to ontologists. The DISO collection includes 15 such ontologies that are widely recognised in the literature. A prominent example is the MITRE Corporation D3FEND cyber-security ontology [30]. Information exchange. Ontologies in this cluster seek to standardise data representa- tions so as to facilitate the exchange (and aggregation) of data across diverse domains, sectors, organisations and systems. The DISO collection contains two authoritative in- formation exchange ontologies. One, IES (Information Exchange Standard) 17 , is a stan- dard for information exchange developed within the UK Government. DISO contains the IES version released in November 2025. The IES plays a critical role in enabling the UKâs vision of a National Digital Twin (NDT). The other information exchange ontology in DISO is JC3IEDM, standardised by NATO and jointly developed under the Multilateral Interoperability Programme (MIP) for the exchange of consultation, command and control (C3) information. Mid-level and Upper-level. In addition to domain/application level ontologies, DISO also contains clusters for mid-level and upper-level ontologies as well. Upper-level on- tologies seek to describe general concepts to embrace and orchestrate domain-level on- tologies. Mid-level ontologies specialise upper-level ontologies for particular (vertical) domains, and thereby play a fundamental bridging role between (general) upper-level ontologies and (specialised) domain-level ontologies. The criticality of mid-level and upper-level ontologies for enabling integration of diverse, heterogeneous knowledge sources is evidenced by the next version of IES, which is being modularised with âies- topâ and âies-coreâ to provide the common foundations for IES domain and sub-domain level ontologies. 2.1 Towards a network of DISO ontologies Ontology alignment [13] is the process of finding correspondences or an alignmentM among the entities (ontology classes, properties or instances) of two or more ontolo- gies. A mapping involving two entities is typically represented as a 4-tuple âše 1 , e 2 , r, câ© where e 1 and e 2 are entities of the ontologiesO 1 andO 2 , respectively, r is a seman- tic relation, typically one ofâ,â,âĄ, and c is a confidence value (usually a number between 0 and 1). An ontology alignment system is a program that, given as input two ontologies, generates an ontology alignmentM S between them. We conducted a bulk ontology alignment exercise using the ontology alignment sys- tem LogMap [27,28] over the DISO ontologies, resulting in the (potential) generation of alignments for 1,653 ontology pairs. Table 2 compares the distributions of alignment sizes with respect to two different sets of DISO ontologies: (i) the full set of 1,653 DISO ontology pairs, and (i) the subset of 105 DISO cyber-security ontology pairs derived from the 15 cyber-security ontologies. We observe that LogMap identifies an intersection between at least 647 ontology pairs (66 about cyber-security), indicating that the DISO ontology network has an important number of connections (i.e., ontol- ogy mappings) across its ontologies as we advanced in Figure 1. The analysis of the bulk ontology alignment served as the basis to select a subset of ontology pairs for a 17 https://informationexchangestandard.org/ 8J. Dilworth et al. Table 2: A comparison of the (partial) distributions of alignment size (measured in entity mapping counts) for two different sets of DISO ontology pairs, as generated by one ontology matching system. 1653 DISO ontology pairs105 DISO cyber-security ontology pairs AlignmentOntologyOntologyCumulativeOntologyOntologyCumulative MappingPair Count Pair Count Pair Count AsPair Count Pair CountPair Count As CountCumulative Proportion ofCumulativeProportion of Total PairsTotal Pairs 0100610060.6139390.37 1 21312190.7417560.53 210713260.8013690.66 3 9014160.8611800.76 44814640.893830.79 52914930.903860.82 6 3115240.926920.88 71715410.932940.90 81415550.941950.90 91315680.950950.90 â„109516531.0101051.0 new ontology alignment track, as described in the next section. Ultimately, the goal is to develop an integrated network of DISO ontologies that contributes to addressing the interoperability challenge in the defence and national security domain. 3 The DISO OAEI track The Ontology Alignment Evaluation Initiative (OAEI) [49] is an annual campaign for the systematic evaluation of ontology alignment systems. The OAEI plays a key role in the benchmarking, comparison and reproducibility of ontology alignment systems. The OAEI includes different matching tasks involving small-sized (e.g., conference [61]), medium-sized (e.g., anatomy [10]), and large (e.g., bio-ml [22]) ontologies. Interoperability in the defence and national security domain has been recognised as a major challenge, and a community-driven initiative such as the OAEI can contribute to the integration of the DISO ontologies. At the same time, the defence and national security domain offers a promising real-world scenario and a set of non-trivial matching tasks for the campaign, as shown in this section. The bulk ontology analysis revealed an important potential intersection among the DISO ontologies. We analysed the ontol- ogy pairs sharing a larger amount of ontology mappings (as suggested by the LogMap system) and ranked them in terms of complexity of the alignment task; that is, ontology pairs where their entities have fewer identical labels were preferred. LogMap was se- lected to drive the task selection as it generates different types of outputs that enabled a fine-grained analysis. Additional information is provided in the GUARD project report (phase 1) [26]. Table 3 shows the top-8 candidate ontology pairs that emerged from Improving Interoperability among Defence and National Security Ontologies9 Table 3: The top-8 ontology pairs that emerged from the LogMap-driven analytic process. âSAâ refers to âsituation awarenessâ; âCAâ refers to âcontext awarenessâ. Task Subdomain Source OntologyTarget OntologyLogMap Alignment NameClusterNameClusterAll Cls Oprop DpropInd cybersecurity UCOcybersecuritySTIXcybersecurity131 434840 STIXcybersecurityD3FEND cybersecurity37 31060 SA JC3IEDMSAmIO!CA219 1200 207 JC3IEDMSABricksmart buildings4960043 JC3IEDMSAFacilitycco-modules44 44000 smart envir ThinkHome smart homesBricksmart buildings33 24017 Bricksmart buildingsSmartEnv smart homes32 25007 CityOWLsmart citiesBricksmart buildings24 16107 the analytic process and form the basis for the new DISO-OAEI track. These ontology pairs cluster into three distinct subdomains: cybersecurity (2 pairs), situation awareness (3 pairs), and smart environments (3 pairs). 3.1 Ontology statistics and provenance Table 4 shows the statistics of the ten selected ontologies for the DISO-OAEI matching tasks. For each ontology, we briefly describe its domain and provenance. UCO: Unified Cyber Ontology UCO 18 is a domain-level ontology assigned to the DISO âcyber-securityâ cluster. It focuses on concepts such as cyber investigations, network defence, threat intelligence, malware analysis, vulnerability research, and offensive op- erations. UCO is an open, community-developed ontology, partly funded by the Linux Foundation, with the ambition to support the cybersecurity domain in its entirety. STIX: Structured Threat Information eXpression. STIX [5] is a domain-level ontology assigned to the DISO âcyber-securityâ cluster. Initially developed by MITRE Corpora- tion, STIX was designed as a language and serialisation format for exchanging cyber threat intelligence. The official STIX specification is managed by the OASIS Cyber Threat Intelligence (CTI) Technical Committee. 19 D3FEND. D3FEND [30] is a domain-level ontology assigned to the DISO âcyber- securityâ cluster. MITRE Corporation, its developer, refers to it as a knowledge graph of cybersecurity countermeasures, and as a framework for cybersecurity operations and strategic decision-making. Development of D3FEND was funded by the National Se- curity Agency (NSA), the U.S. Cyber Warfare Directorate, and the U.S. Office of the Under Secretary of Defense for Research and Engineering. JC3IEDM: Joint Consultation, Command and Control Information Exchange Data Model. JC3IEDM [43], standardised under NATO as STANAG 5525, is a domain-level ontology assigned to the DISO clusters âinformation exchangeâ and âsituation aware- nessâ [58]. The aim of the JC3IEDM is to support the exchange of consultation, com- mand and control (C3) information, where heterogeneous command and control in- 18 https://unifiedcyberontology.org/ 19 https://oasis-open.github.io/cti-documentation/ 10J. Dilworth et al. Table 4: Quantitative descriptions of the 10 ontologies appearing in the DISO-OAEI matching tasks. The metrics are as reported by the ProtĂ©gĂ© (https://protege.stanford.edu/). MetricUCO STIX D3FEND JC3IEDM mIO!Brick Facility ThinkHome SmartEnv CityOWL Axioms11,552 3,90332,83226,622 7,727 117,8117,0838,1224,1467,387 Class count429883,4952,9226242,3058591,109161516 Object property count19268205619364131146406242522 Data property count58138642313310801230350162 Individual count50742,3404,0885208,55681541717 Annotation property count4363522416540224442 Class axioms SubClassOf444 12705,3912,5676272,4949252,3444011,250 EquivalentClasses000272361052528262 DisjointClasses818806215231232136 Object property axioms SubObjectPropertyOf11025009219846116448 EquivalentObjectProperties000004001313 InverseObjectProperties90411167096348807 ObjectPropertyDomain108925033440127350177517 ObjectPropertyRange1817145033372128340173511 Data property axioms SubDataPropertyOf401420926012139 EquivalentDataProperties0000000000 DataPropertyDomain44482313304288248162 DataPropertyRange579385173132842827743159 Individual axioms ClassAssertion55801,1944,088904 18,88283532925 formation systems (C2IS) are sharing military situational awareness data. This model has been further developed under the Multilateral Interoperability Programme (MIP) to deliver the MIP Information Model. 20 mIO!: A Context Ontology for Mobile Environments. mIO! [50] is a domain-level on- tology assigned to DISO cluster âcontext awarenessâ. It is a context ontology network whose aim is to model context-related knowledge that allows mobile applications to adapt their behaviour based on user context. mIO! was developed by the Ontology En- gineering Group at University of Madrid. 21 Brick: A Uniform Metadata Schema for Buildings. Brick 22 is a domain-level ontology assigned to DISO cluster âsmart buildingsâ (âsmart environmentsâ). It is an open-source project whose aim is to standardise descriptions of physical, logical and virtual aspects of buildings, and the relationships between them. The Brick consortium comprises com- mercial and academic members. Facility. Facility is a CCO (Common Core Ontology) 23 mid-level ontology assigned to DISO cluster âcco-modulesâ, within cluster âmid-levelâ. This mid-level ontology was designed to represent facilities (such as buildings, campuses, etc.) that serve some spe- cific purpose and which are common in multiple domains. 20 The MIP Information Model (2025): https://w.mimworld.org/ 21 https://oeg.fi.upm.es/index.php/en/ontologies/index.html 22 https://brickschema.org/ 23 https://w.commoncoreontologies.org Improving Interoperability among Defence and National Security Ontologies11 ThinkHome: Smart Home Ontology for Human Activity Recognition. ThinkHome [51] is a domain-level ontology assigned to DISO cluster âsmart homesâ (âsmart environ- mentsâ). It is an ontology intended to support smart systems, especially those with a focus on home control, and particularly on energy efficiency. It consists of five com- ponent ontologies that address individual aspects of the target application domain: ac- tors, buildings, energy resources, processes and weather. ThinkHome was developed at Technical University Vienna. 24 SmartEnv: Smart Home Environments. SmartEnv [1,34] is a domain-level ontology as- signed to DISO cluster âsmart homesâ (âsmart environmentsâ). SmartEnv, initially re- ferred to as âE-care@homeâ in the literature [2], is a network of 8 component ontologies modelling various aspects of smart homes: temporal, spatial, objects, events, agents, observations, etc. Its notion of âsmartâ refers to environments (such as homes) that are monitored by sensors of various kinds. CityOWL. CityOWL is a domain-level ontology assigned to DISO cluster âsmart citiesâ (âsmart environmentsâ). It is an OWL rendering of a standard schema defined by the Open Geospatial Consortium called CityGML 25 . The standard defines a conceptual model for representing virtual 3D models (aka digital twins) of cities. The CityOWL ontology obtained for DISO was acquired from a URL associated with the Computer Science Laboratory for Image and Information Systems at the University of Lyon. 26 3.2 Alignment evaluation We executed several state-of-the-art systems on the DISO-OAEI matching tasks with two main objectives: (i) to assess the complexity of the tasks, and (i) to generate silver-standard alignments for the track. We selected the following systems: AML [15], BertMap [21] (with variations using different BERT-based pre-trained models), BertMapLt, LogMap [27], LogMapLt, LogMapLLM [37], Matcha [16], ALOD2Vec [48], ATMatcher [24], Fine-TOM [33] and KGMatcher [14]. Table 5 presents the mappings computed by the ontology alignment systems in each of the DISO-OAEI tasks. Note that LogMapLLM, ALOD2Vec, ATMatcher, Fine-TOM, and KGMatcher were not able to produce map- pings on all matching tasks. It can also be observed that the systems generated highly disparate sets of mappings in the DISO-OAEI tasks. For example, in the JC3IEDM- mIO! task, LogMap produces 234 mappings while AML generates 34. Consensus alignments. We have computed consensus alignments of vote 2 and 3 (i.e., mappings suggested by at least two or three systems, respectively). Note that, when there are several systems of the same family (i.e., systems participating with several variants), their (voted) mappings are only counted once in order to reduce bias. Table 5 shows the number of consensus mappings voted by 2, 3 and all systems (Con-2, Con-3 and Con-all, respectively) and the number of system families contributing to the consen- sus (#SF). For example, in the JC3IEDM-mIO! task, 77 mappings were suggested by at least 2 system families, while only 11 mappings were recommended by all 8 families. 24 https://w.auto.tuwien.ac.at/index.php/projectsites/ 155-thinkhome 25 https://w.ogc.org/standards/citygml/ 26 https://liris.cnrs.fr/ 12J. Dilworth et al. Table 5: Number of system and consensus mappings for the selected matching tasks. #SF: number of system families contributing to the consensus. Con-x: consensus map- pings with âxâ votes. Con-all: consensus mapping supported by all system families. Matching task System mappingsConsensus mappings AMLBertMapBertMapLtLogMapLogMapLtMatcha#SFCon-2Con-3Con-all UCO-STIX43212113117769241713921 STIX-D3FEND27171837281026453217 JC3IEDM-mIO!342226234371678774611 JC3IEDM-Brick4523235710577641082722 JC3IEDM-Facility432829443743871412 ThinkHome-Brick1121317551413124955816 Brick-SmartEnv32122239523224482822 CityOWL-Brick38152227584024642823 Table 6: Manual validation of unique and consensus mappings. For each system family, we provide the number of unique mappings and the ratio of correct ones in brackets. For Matcha, the ratio represents an approximation given the large number of generated mappings. Others aggregates ALOD2VEc, ATMatcher, Fine-TOM and KGMatcher. . Matching Task Unique system mappings by familyConsensus-2 mappingsSilver AMLBertMapLogMapMatchaOthers# Total% Correct# Total UCO-STIX1 (0%)1 (0%)43 (84%)516 (10%)0 (0%)17191%194 STIX-D3FEND2 (0%)0 (0%)10 (20%)63 (10%)0 (0%)4576%39 JC3IEDM-mIO!3 (0%)0 (0%)199 (97%)113 (7%)65 (9%)7787%274 JC3IEDM-Brick3 (0%)2 (50%)38 (5%)662 (2%)0 (0%)10867%75 JC3IEDM-Facility6 (0%)3 (33%)5 (60%)0 (0%)34 (0%)4494%44 ThinkHome-Brick45 (13%)0 (0%)80 (47%)299 (6%)0 (0%)9592%134 Brick-SmartEnv3 (0%)0 (0%)36 (31%)274 (10%)0 (0%)4877%53 CityOWL-Brick6 (0%)0 (0%)21 (57%)339 (5%)0 (0%)6483%68 Manual assessment. Consensus alignment can serve as a reference for the agreement across systems; they are, however, not complete and may also include errors, as it only requires two systems to agree on an incorrect mapping [20]. Hence, we have performed a manual revision of the Con-2 consensus mappings and the mappings uniquely gen- erated by a system or system family. Unique mappings may include potentially valid correspondences that were not identified by any other system family. Table 6 presents the results of the manual validation. Consensus mappings are typically precise, with 94% of the mappings in JC3IEDM-Facility classified as correct. Con-2 was less accu- rate for the JC3IEDM-Brick task. The quality of unique mappings varied across systems and tasks. For example, 80% and 13% of the unique mappings identified by LogMap and AML in ThinkHome-Brick, respectively, were annotated as correct. The manual re- vision led to the creation of a silver-standard for the DISO-OAEI tracks by merging the correct unique mappings and the correct Con-2 mappings. The manual validation was performed primarily by the authors. However, a broader validation involving multiple domain experts, together with an inter-annotator agreement analysis, is planned in the context of the OAEI 2026 campaign. The quality of the silver reference will also im- prove as more systems contribute mappings, and it will be updated with results from the OAEI participants. Improving Interoperability among Defence and National Security Ontologies13 0.00.20.40.60.81.0 MDS-1 0.0 0.2 0.4 0.6 0.8 1.0 MDS-2 UCOSTIX BertMap BertMapLt CySecBertMap SecureBertMap LogMap LogMapLLM AML LogMapLt Matcha BertMap Family LogMap Family Con-2 Con-3 s-ref 0.00.20.40.60.81.0 MDS-1 0.0 0.2 0.4 0.6 0.8 1.0 MDS-2 STIXD3FEND BertMap BertMapLt CySecBertMap KGMatcher SecureBertMap ALOD2Vec AML LogMap LogMapLLM LogMapLt Matcha BertMap Family LogMap Family Con-2 Con-3 s-ref 0.00.20.40.60.81.0 MDS-1 0.0 0.2 0.4 0.6 0.8 1.0 MDS-2 JC3IEDMmIO BertMap CySecBertMap SecureBertMap LogMap LogMapLLM ALOD2Vec AML ATMatcher BertMapLt Fine-TOM KGMatcher LogMapLt Matcha BertMap Family LogMap Family Con-2 Con-3 s-ref 0.00.20.40.60.81.0 MDS-1 0.0 0.2 0.4 0.6 0.8 1.0 MDS-2 ThinkHomeBrick BertMap CySecBertMap SecureBertMap AML BertMapLt LogMap LogMapLt Matcha BertMap Family LogMap Family Con-2 Con-3 s-ref SystemFamily unionConsensusSilver reference Fig. 2: Visual representation of the Jaccard distances among mapping sets. Mapping comparison. Table 5 and the consensus mapping extraction have revealed a significant disparity among the evaluated systems, which is further illustrated in Fig- ure 2. This confirms that the selected matching task involves the discovery of non-trivial mappings. The figure presents two-dimensional scatterplots of the Jaccard distances between the compared mapping sets. Proximity to the consensus mapping set is not necessarily an indicator of better quality; rather, it reflects a higher agreement with the consensus. For example, in JC3IEDM-mIO!, LogMapLt is close to the Con-3 consen- sus but far from s-ref (the manually-curated silver-standard reference). System families are also visualised, representing the aggregation of all mappings produced by the cor- responding system variants. We emphasise that the silver-standard is derived from the outputs of the participating systems, its recall is inherently bounded by what these sys- tems can collectively discover, and it will favour systems that resemble those used in its construction. Thus, it should be regarded as a partial reference rather than a complete gold standard. For the same reason, we prefer not to report the actual precision and recall values of the participating systems with respect to the silver standard, but rather the proximity among the relevant mapping sets, as shown in Figure 2. 14J. Dilworth et al. Table 7: Unsatisfiable classes led by system, consensus and silver-standard mappings, computed over the whole merged ontology with the OWL 2 El reasoner ELK [32], marked byâ„ (lower-bound), or a complete OWL 2 DL reasoner (JFact [56], otherwise HermiT [19]). incons. denotes a merged TBox detected as inconsistent by ELK. Matching task System mappingsAggregated mappings AMLBertMapBertMapLtLogMapLogMapLtMatchaCon-2Silver UCO-STIX122 (24%)110 (21%)110 (21%)134 (26%)131 (25%)127 (24%)122 (24%)133 (25%) STIX-D3FEND2448 (68%)2441 (68%)2444 (68%)2516 (70%)2516 (70%)2509 (70%)2509 (70%)2506 (70%) JC3IEDM-mIO!00000000 JC3IEDM-Brick00000000 JC3IEDM-Facility00000000 ThinkHome-Brick1123 (33%)0â„49 (1.4%)01375 (40%)â„652 (19%)907 (27%)874 (26%) Brick-SmartEnvâ„120â„110â„844 (34%)incons.â„844 (34%)â„842 (34%) CityOWL-Brick26 (0.9%)020 (0.7%)020 (0.7%)279 (9.9%)30 (1.1%)21 (0.7%) Ontology compatibility. Reasoning with the input ontologies and the mappings may lead to logical errors (e.g., unsatisfiabilities). An alignment M is incoherent if O 1 âȘ O 2 âȘM |= Aâ â„ (for any class A). Table 7 shows the logical errors led by system- computed mappings and the generated consensus and silver-standard alignments. It is worth noting that even systems implementing (approximate) reasoning techniques like AML and LogMap [54] lead to a large number of unsatisfiabilities (e.g., UCO-STIX and STIX-D3FEND). This highlights a modelling disagreement among the DISO-OAEI ontologies, bringing an interesting challenge to the OAEI from the logical and reason- ing point of view. The silver-standard alignment, which involved manual curation, also leads to unsatisfiable classes, emphasising the need for reasoning techniques within the ontology alignment pipeline. For the DISO-OAEI 2026 track, the silver standard has been treated to fix logical errors and annotate incoherent mappings. 4 The DISO ecosystem Resources are made available through a small ecosystem of three repositories, as de- picted in Figure 3. In particular, the collection of defence and national security on- tologies is curated in the diso repository 27 and mirrored to Zenodo 28 upon every major release. Since some of the ontologies are packaged as networked component ontologies, we ship both a canonical distribution and a compact distribution. The compact distribu- tion merges networked ontologies into a single file for the convenience of downstream consumers. For instance, the compact distribution is consumed by our alignment ex- traction pipeline diso-mappings 29 , which is the second repository in our ecosystem. It allows contributors to easily download the compact distribution, register new matching systems, and iteratively improve the reference alignment. diso-mappings is a Python pipeline that consumes the DISO compact distribution and produces pairwise align- ments between selected ontology pairs. It uses a configurable set of ontology matching 27 https://github.com/city-artificial-intelligence/diso 28 https://doi.org/10.5281/zenodo.20059506 29 https://github.com/city-artificial-intelligence/diso-mappings Improving Interoperability among Defence and National Security Ontologies15 DISO -mappings Knowledge Curators (Contributors) Maintain, improve and extend the network of ontologies. Manually review, adjust, and verify the consensus alignment. Maintainers & Domain Experts matchers mappings consensus DISO Network of Ontologies Reference Alignment DISO -OAEI Participants Submit ontology matching systems for evaluation. LogMap[Lt], BERTMap[Lt], AML benchmarking Future Benchmarks and Leaderboard Fig. 3: The DISO ecosystem. The diso ontology collection aims to be extended through a collective effort by contributors and released through Zenodo. The diso-mapping pipeline runs a configurable, extendable set of matchers and aggregates their outputs into a family-based consensus alignment. The alignment is reviewed by maintainers and domain experts, and serves as a partial silver-standard reference alignment, form- ing the foundation for the diso-oaei track, where OAEI systems will be benchmarked. systems (AML, LogMap, LogMapLt, BERTMap, and BERTMapLt by default; with an extendable Matcher base class that enables custom external matchers). The pipeline aggregates the per-system alignments into a consensus alignment via a family-based voting mechanism. This alignment is then manually verified, adapted and used as the basis for a partial silver-standard reference alignment. The computed alignments can easily be reproduced by following the steps within the repositoryâs make workflow. diso-oaei 30 is the home for the new DISO-OAEI track introduced in the OAEI 2026 edition. It contains the datasets, evaluation metrics, participation instructions, and will also host the OAEI benchmarking results. 5 Conclusions and sustainability of the resources In this work, we have curated and described a collection of publicly available OWL on- tologies covering the defence and national security domain, and proposed a new OAEI track for evaluating ontology matching systems on this collection. Our analysis iden- tified eight candidate ontology pairs, spanning three distinct subdomains, supporting 30 https://github.com/city-artificial-intelligence/diso-oaei 16J. Dilworth et al. non-trivial ontology matching tasks across classes, properties, and instances. We re- leased three resources in support of a rich and diverse track: diso, the curated ontology collection for defence, intelligence and national security; diso-mappings, a reproducible pipeline that aggregates the output of several state-of-the-art matchers into a consensus alignment via family-based voting; and diso-oaei, the home page for the new OAEI 2026 track. The consensus alignment was manually reviewed (by the authors), leading to a silver-standard reference alignment. This effort has been conducted as part of the GUARD project, funded by The Tur- ing Defence & Security Grand Challenge. We are actively collaborating with experts from the UK Government intelligence community, the NDTP (National Digital Twin Programme), and the Dstl (Defence Science and Technology Laboratory), who will support the development and maintenance of the DISO ecosystem. The DISO ecosys- tems aim at serving both the broader research community and the UKâs national needs on defence and national security with regard to the improvement of the interoperability of the relevant domain ontologies. The defence and national security domain presents distinctive challenges for on- tology matching: source ontologies span heterogeneous granularities (e.g., upper-level, mid-level, temporal), numerous subdomains, and gold standard ground-truth coverage is limited. The OAEI community-driven evaluation will help to inform the future direc- tion of research in this domain. The OAEI evaluation and the collaboration with domain experts will also support the extensions and improvement of the DISO ecosystem: on- tologies, matching tasks, silver-standards, and, ultimately, the creation of an integrated network of DISO ontologies. For the OAEI 2026, we also plan to include automated benchmarking procedures, an active leaderboard, and a candidate mapping ranking task as in the bio-ml track [22] to attract the participation of machine learning systems. DISO is not the first initiative to gather domain-specific ontologies. A prominent example is BioPortal [45], which has become the reference repository for biomedical ontologies. Similarly, the National Center for Ontological Research (NCOR) is devel- oping a National Security Ontology Foundry, 31 with which we intend to explore po- tential collaborations. As DISO evolves, we will also explore adopting the OntoPortal technology to host and manage its ontologies [29]. 32 Acknowledgements. This research was supported by Turing Innovations Limited and The Alan Turing Instituteâs Defence and Security Programme via the project GUARD. It was also supported by FCT through the fellowship https://doi.org/10.54499/ 2022.10557.BD (Pedro Cotovio), the LASIGE Research Unit, ref. UID/00408/2025 and the CancerScan project, which received funding from the European Unionâs Hori- zon Europe Research and Innovation Action (EIC Pathfinder Open) under grant agree- ment No. 101186829. Views and opinions expressed are, however, those of the author(s) only and do not necessarily reflect those of The Alan Turing Institute, the European Union or the European Innovation Council and SMEs Executive Agency. Neither the European Union nor the granting authority can be held responsible for them. 31 https://ncor-network.org/ 32 https://ontoportal.org/ Improving Interoperability among Defence and National Security Ontologies17 Declaration of use of Generative AI. AI tools were used solely for grammar correc- tion and minor language edits. They did not contribute to content creation, idea develop- ment, or substantive rewriting. The research design, experiments, analysis, and writing were carried out entirely by the authors. References 1. Alirezaie, M., Hammar, K., Blomqvist, E.: SmartEnv as a network of ontology patterns. Semantic Web 9(6), 903â918 (2018). https://doi.org/10.3233/SW-180303 2. Alirezaie, M., Renoux, J., Köckemann, U., Kristoffersson, A., Karlsson, L., Blomqvist, E., Tsiftes, N., Voigt, T., Loutfi, A.: An Ontology-based Context-aware System for Smart Homes: E-care@home. Sensors 17(7) (2017). https://doi.org/10.3390/ S17071586 3. Augusto, J.C., McCullagh, P.J.: Ambient Intelligence: Concepts and applications. Comput. Sci. Inf. Syst. 4(1), 1â27 (2007). https://doi.org/10.2298/CSIS0701001A 4. Bailey, I., Beverley, J., Blackmore, H., Cola, A., Cripps, P., Colle, G.D., Donato, F., Hicks, A., Limbaugh, D., Milivinti, E., Partridge, C., Rafferty, R., Smith, B.: Comparing Informa- tion Exchange Standard and Basic Formal Ontology Design Patterns. In: Proceedings of the Joint Ontology Workshops (JOWO). CEUR Workshop Proceedings, CEUR-WS.org (2025), https://ceur-ws.org/Vol-4176/foust-6.pdf 5. Barnum, S.: Standardizing Cyber Threat Intelligence Information with the Structured Threat Information eXpression (STIX). Tech. rep., MITRE Corporation (2014), https: //stixproject.github.io/getting-started/whitepaper/ 6. Chen, H., Perich, F., Finin, T.W., Joshi, A.: SOUPA: Standard Ontology for Ubiquitous and Pervasive Applications. In: 1st Annual International Conference on Mobile and Ubiquitous Systems (MobiQuitous). p. 258â267. IEEE Computer Society (2004). https://doi. org/10.1109/MOBIQ.2004.1331732 7. Chen, J., Lu, Y., Zhang, Y., Huang, F., Qin, J.: A management knowledge graph approach for critical infrastructure protection: Ontology design, information extraction and relation prediction. Int. J. Crit. Infrastructure Prot. 43, 100634 (2023). https://doi.org/10. 1016/J.IJCIP.2023.100634 8. Costa, D., Collins, M., Perl, S.J., Albrethsen, M., Silowash, G., Spooner, D.: An ontology for insider threat indicators: Development and application. In: Proceedings of the Ninth Con- ference on Semantic Technology for Intelligence, Defense, and Security. CEUR Workshop Proceedings, vol. 1304, p. 48â53. CEUR-WS.org (2014), https://ceur-ws.org/ Vol-1304/STIDS2014_T07_CostaEtAl.pdf 9. Cuenca Grau, B., Horrocks, I., Motik, B., Parsia, B., Patel-Schneider, P.F., Sattler, U.: OWL 2: The next step for OWL. J. Web Semant. 6(4), 309â322 (2008), https://doi.org/ 10.1016/j.websem.2008.05.001 10. Dragisic, Z., Ivanova, V., Li, H., Lambrix, P.: Experiences from the anatomy track in the ontology alignment evaluation initiative. J. Biomed. Semant. 8(1), 56:1â56:28 (2017), https://doi.org/10.1186/s13326-017-0166-5 11. Dragos, V.: Ontology modeling for intelligence: the ONTO-CIF model. In: Advances in Knowledge-Based and Intelligent Information and Engineering Systems - 16th Annual KES Conference. Frontiers in Artificial Intelligence and Applications, vol. 243, p. 1543â1552. IOS Press (2012). https://doi.org/10.3233/978-1-61499-105-2-1543 12. Endsley, M.R.: Toward a Theory of Situation Awareness in Dynamic Systems. Human Fac- tors 37(1), 32â64 (1995). https://doi.org/10.1518/001872095779049543 13. Euzenat, J., Shvaiko, P.: Ontology Matching, Second Edition. Springer (2013) 18J. Dilworth et al. 14. Fallatah, O., Zhang, Z., Hopfgartner, F.: Kgmatcher results for OAEI 2021. In: Shvaiko, P., Euzenat, J., JimĂ©nez-Ruiz, E., Hassanzadeh, O., Trojahn, C. (eds.) Proceedings of the 16th International Workshop on Ontology Matching co-located with the 20th International Semantic Web Conference (ISWC 2021), Virtual conference, October 25, 2021. p. 160â 166. CEUR Workshop Proceedings, CEUR-WS.org (2021), https://ceur-ws.org/ Vol-3063/oaei21_paper8.pdf 15. Faria, D., Santos, E., Balasubramani, B.S., Silva, M.C., Couto, F.M., Pesquita, C.: Agree- mentMakerLight. Semantic Web 16(2), SWâ233304 (2025) 16. Faria, D., Silva, M.C., Cotovio, P., Ferraz, L., Balbi, L., Pesquita, C.: Results for Matcha and Matcha-DL in OAEI 2023. In: Proceedings of the 18th International Workshop on On- tology Matching co-located with the 22nd International Semantic Web Conference (ISWC 2023), Athens, Greece, November 7, 2023. CEUR Workshop Proceedings, vol. 3591, p. 164â169. CEUR-WS.org (2023), https://ceur-ws.org/Vol-3591/oaei23_ paper6.pdf 17. dâAvila Garcez, A., Lamb, L.C.: Neurosymbolic AI: the 3rd wave. Artif. Intell. Rev. 56(11), 12387â12406 (2023). https://doi.org/10.1007/S10462-023-10448-W 18. Garijo, D., Poveda-VillalĂłn, M.: Best Practices for Implementing FAIR Vocabularies and Ontologies on the Web (2020), https://arxiv.org/pdf/2003.13084 19. Glimm, B., Horrocks, I., Motik, B., Stoilos, G., Wang, Z.: HermiT: An OWL 2 Rea- soner. J. Autom. Reason. 53(3), 245â269 (2014). https://doi.org/10.1007/ S10817-014-9305-1 20. Harrow, I., JimĂ©nez-Ruiz, E., Splendiani, A., Romacker, M., Woollard, P., Markel, S., Alam- Faruque, Y., Koch, M., Malone, J., Waaler, A.: Matching disease and phenotype ontologies in the ontology alignment evaluation initiative. J. Biomed. Semant. 8(1), 55:1â55:13 (2017). https://doi.org/10.1186/S13326-017-0162-9 21. He, Y., Chen, J., Antonyrajah, D., Horrocks, I.: BERTMap: A BERT-Based Ontology Align- ment System. In: Thirty-Sixth AAAI Conference on Artificial Intelligence. p. 5684â5691. AAAI Press (2022), https://doi.org/10.1609/aaai.v36i5.20510 22. He, Y., Chen, J., Dong, H., JimĂ©nez-Ruiz, E., Hadian, A., Horrocks, I.: Machine Learning- Friendly Biomedical Datasets for Equivalence and Subsumption Ontology Matching. In: 21st International Semantic Web Conference. Lecture Notes in Computer Science, vol. 13489, p. 575â591. Springer (2022), https://doi.org/10.1007/978-3-031-19433-7_ 33 23. Herron, D., JimĂ©nez-Ruiz, E., Weyde, T.: On the Potential of Logic and Reasoning in Neu- rosymbolic Systems using OWL-based Knowledge Graphs. Neurosymbolic Artificial Intel- ligence 1 (April 2025). https://doi.org/10.1177/29498732251320043 24. Hertling, S., Paulheim, H.: ATBox results for OAEI 2022. In: Shvaiko, P., Euzenat, J., JimĂ©nez-Ruiz, E., Hassanzadeh, O., Trojahn, C. (eds.) Proceedings of the 17th In- ternational Workshop on Ontology Matching (OM). p. 153â157. CEUR Workshop Proceedings, CEUR-WS.org (2022), https://ceur-ws.org/Vol-3324/oaei22_ paper4.pdf 25. Hogan, A., Blomqvist, E., Cochez, M., dâAmato, C., de Melo, G., Gutierrez, C., Kirrane, S., Gayo, J.E.L., Navigli, R., Neumaier, S., Ngomo, A.N., Polleres, A., Rashid, S.M., Rula, A., Schmelzeisen, L., Sequeda, J.F., Staab, S., Zimmermann, A.: Knowledge graphs. ACM Comput. Surv. 54(4), 71:1â71:37 (2022), https://doi.org/10.1145/3447772 26. JimĂ©nez-Ruiz, E., Herron, D., Dilworth, J.: Phase 1 Report (GUARD / G2049a): Ensuring Interoperable and Trustworthy Knowledge Graphs for Defence and National Security AI (May 2026), https://openaccess.city.ac.uk/id/eprint/37459/ 27. JimĂ©nez-Ruiz, E., Cuenca Grau, B.: LogMap: Logic-Based and Scalable Ontology Matching. In: The Semantic Web - ISWC 2011 - 10th International Semantic Web Conference. Lecture Improving Interoperability among Defence and National Security Ontologies19 Notes in Computer Science, vol. 7031, p. 273â288. Springer (2011), https://doi. org/10.1007/978-3-642-25073-6_18 28. JimĂ©nez-Ruiz, E., Cuenca Grau, B., Zhou, Y., Horrocks, I.: Large-scale interactive ontology matching: Algorithms and implementation. In: ECAI - 20th European Conference on Artifi- cial Intelligence. Frontiers in Artificial Intelligence and Applications, vol. 242, p. 444â449. IOS Press (2012), https://doi.org/10.3233/978-1-61499-098-7-444 29. Jonquet, C., Graybeal, J.B., Bouazzouni, S., Dorf, M., Fiore, N., Kechagioglou, X., Red- mond, T., Rosati, I., Skrenchuk, A., Vendetti, J., Musen, M.A.: Ontology repositories and semantic artefact catalogues with the ontoportal technology. In: 22nd International Seman- tic Web Conference (ISWC). Lecture Notes in Computer Science, vol. 14266, p. 38â58. Springer (2023). https://doi.org/10.1007/978-3-031-47243-5_3 30. Kaloroumakis, P.E., Smith, M.J.: Toward a Knowledge Graph of Cybersecurity Coun- termeasures. Tech. rep., MITRE Corporation (2021), https://d3fend.mitre.org/ resources/D3FEND.pdf 31. Karabulut, E., Pileggi, S.F., Groth, P., Degeler, V.: Ontologies in digital twins: A systematic literature review. Future Gener. Comput. Syst. 153, 442â456 (2024). https://doi.org/ 10.1016/J.FUTURE.2023.12.013 32. Kazakov, Y., Krötzsch, M., SimancĂk, F.: The incredible ELK - from polynomial procedures to efficient reasoning withEL ontologies. J. Autom. Reason. 53(1), 1â61 (2014). https: //doi.org/10.1007/S10817-013-9296-3 33. Knorr, L., Portisch, J.: Fine-tom matcher results for OAEI 2021. In: Shvaiko, P., Euzenat, J., JimĂ©nez-Ruiz, E., Hassanzadeh, O., Trojahn, C. (eds.) Proceedings of the 16th International Workshop on Ontology Matching co-located with the 20th International Semantic Web Con- ference (ISWC 2021), Virtual conference, October 25, 2021. p. 144â151. CEUR Workshop Proceedings, CEUR-WS.org (2021), https://ceur-ws.org/Vol-3063/oaei21_ paper6.pdf 34. Köckemann, U., Alirezaie, M., Renoux, J., Tsiftes, N., Ahmed, M.U., Morberg, D., LindĂ©n, M., Loutfi, A.: Open-Source Data Collection and Data Sets for Activity Recognition in Smart Homes. Sensors 20(3) (2020). https://doi.org/10.3390/S20030879 35. Leblanc, E., Nguyen, D., Balduccini, M., Regli, W., Kopena, J., Wambold, T.: Military on- tologies for information dissemination at the tactical edge. In: Joint Ontology Workshops. CEUR (2015) 36. Lourenço, B., AdĂŁo, P., Ferreira, J.F., Marques, M.M., Vaz, C.: Structuring Security: A Sur- vey of Cybersecurity Ontologies, Semantic Log Processing, and LLMs Application. CoRR abs/2510.16610 (2025). https://doi.org/10.48550/ARXIV.2510.16610 37. Lushnei, S., Shumskyi, D., Shykula, S., JimĂ©nez-Ruiz, E., dâAvila Garcez, A.: Large Lan- guage Models as Oracles for Ontology Alignment. In: 19th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2026). Association for Computational Linguists (2026) 38. Maryasin, O.: Home Automation System Ontology for Digital Building Twin. In: 2019 XXI International Conference Complex Systems: Control and Modeling Problems (CSCMP). p. 70â74 (2019). https://doi.org/10.1109/CSCMP45713.2019.8976546 39. Matthews, M., Shattuck, L., Grham, S., Weeks, J., Endsley, M., Strater, L.: Situation Awareness for Military Ground Forces: Current Issues and Perspectives. Human Factors 45(4),351â355(2001), https://w.researchgate.net/publication/ 273594666_Situation_Awareness_for_Military_Ground_Forces_ Current_Issues_and_Perspectives 40. Meriah, I., Rabai, L.B.A., KhĂ©dri, R.: Building a comprehensive and multi-dimensional in- formation security ontology: elicitation process and OWL implementation. Knowl. Inf. Syst. 67(1), 167â195 (2025). https://doi.org/10.1007/S10115-024-02308-Y 20J. Dilworth et al. 41. Morosoff, P., Rudnicki, R., Bryant, J., Farrell, R., Smith, B.: Joint Doctrine Ontology: A Benchmark for Military Information Systems Interoperability. In: 10th Conference on Semantic Technology for Intelligence, Defense, and Security. CEUR Workshop Proceed- ings, vol. 1523. CEUR-WS.org (2015), https://ceur-ws.org/Vol-1523/STIDS_ 2015_T01_Morosoff_etal.pdf 42. Mundie, D.A., Ruefle, R., Dorofee, A.J., Perl, S.J., McCloud, J., Collins, M.: An incident management ontology. In: Proceedings of the Ninth Conference on Semantic Technology for Intelligence, Defense, and Security. CEUR Workshop Proceedings, vol. 1304, p. 62â 71. CEUR-WS.org (2014), https://ceur-ws.org/Vol-1304/STIDS2014_T09_ MundieEtAl.pdf 43. NATO/MIP: THE JOINT C3 INFORMATION EXCHANGE DATA MODEL OVERVIEW. Tech. rep., NATO/MIP (2007), https://brainsciencesjournal.org/2007. JC3IEDM.pdf 44. Naveed, H., Khan, A.U., Qiu, S., Saqib, M., Anwar, S., Usman, M., Akhtar, N., Barnes, N., Mian, A.: A comprehensive overview of large language models. ACM Trans. Intell. Syst. Technol. 16(5) (Aug 2025). https://doi.org/10.1145/3744746 45. Noy, N.F., Shah, N.H., Whetzel, P.L., Dai, B., Dorf, M., Griffith, N., Jonquet, C., Rubin, D.L., Storey, M.D., Chute, C.G., Musen, M.A.: BioPortal: ontologies and integrated data resources at the click of a mouse. Nucleic Acids Res. 37(Web-Server-Issue), 170â173 (2009), https://doi.org/10.1093/nar/gkp440 46. Obrst, L., Chase, P., Markeloff, R.: Developing an Ontology of the Cyber Security Do- main. In: Proceedings of the Seventh International Conference on Semantic Technologies for Intelligence, Defense, and Security. CEUR Workshop Proceedings, vol. 966, p. 49â 56. CEUR-WS.org (2012), https://ceur-ws.org/Vol-966/STIDS2012_T06_ ObrstEtAl_CyberOntology.pdf 47. Oltramari, A., Cranor, L.F., Walls, R.J., McDaniel, P.D.: Building an Ontology of Cy- ber Security. In: Proceedings of the Ninth Conference on Semantic Technology for In- telligence, Defense, and Security. CEUR Workshop Proceedings, vol. 1304, p. 54â 61. CEUR-WS.org (2014), https://ceur-ws.org/Vol-1304/STIDS2014_T08_ OltramariEtAl.pdf 48. Portisch, J., Paulheim, H.: ALOD2Vec matcher results for OAEI 2021. In: Proceedings of the 16th International Workshop on Ontology Matching co-located with the 20th International Semantic Web Conference (ISWC). p. 117â123. CEUR Workshop Proceedings, CEUR- WS.org (2021), https://ceur-ws.org/Vol-3063/oaei21_paper2.pdf 49. Pour, M.A.N., Blomqvist, E., Cotovio, P.G., et al.: Results of the Ontology Alignment Eval- uation Initiative 2025. In: 20th International Workshop on Ontology Matching (OM). p. 105â139 (2025) 50. Poveda-VillalĂłn, M., SuĂĄrez-Figueroa, M.C., Garcia-Castro, R., GĂłmez-PĂ©rez, A.: A Con- text Ontology for Mobile Environments. In: Proceedings of the Second Workshop on Con- text, Information and Ontologies, CIAO@EKAW. CEUR Workshop Proceedings, CEUR- WS.org (2010), https://ceur-ws.org/Vol-626/regular3.pdf 51. Reinisch, C., Kofler, M.J., Iglesias, F., Kastner, W.: ThinkHome Energy Efficiency in Future Smart Homes. EURASIP J. Embed. Syst. 2011 (2011). https://doi.org/10.1155/ 2011/104617 52. Rivadeneira, W., GĂłmez, O.S.: Cybersecurity Ontologies: A Systematic Literature Review. ReCIBE, Revista electrĂłnica de ComputaciĂłn, InformĂĄtica, BiomĂ©dica y ElectrĂłnica 9, C1â C18 (2021). https://doi.org/10.32870/recibe.v9i2.181 53. Sikos, L.F.: Cybersecurity Knowledge Graphs. Knowl. Inf. Syst. 65(9), 3511â3531 (2023). https://doi.org/10.1007/S10115-023-01860-3 Improving Interoperability among Defence and National Security Ontologies21 54. Solimando, A., JimĂ©nez-Ruiz, E., Guerrini, G.: Minimizing conservativity violations in on- tology alignments: algorithms and evaluation. Knowl. Inf. Syst. 51(3), 775â819 (2017). https://doi.org/10.1007/S10115-016-0983-3 55. Stavropoulos, T.G., Vrakas, D., Vlachava, D., Bassiliades, N.: BOnSAI: A Smart Building Ontology for Ambient Intelligence. In: 2nd International Conference on Web Intelligence, Mining and Semantics, WIMS. p. 30:1â30:12. ACM (2012). https://doi.org/10. 1145/2254129.2254166 56. Tsarkov, D., Horrocks, I.: Fact++ description logic reasoner: System description. In: Fur- bach, U., Shankar, N. (eds.) 3rd International Joint Conference on Automated Reasoning (IJCAR). Lecture Notes in Computer Science, vol. 4130, p. 292â297. Springer (2006). https://doi.org/10.1007/11814771_26 57. Valero, J.M.J., MartĂnez, A.L., SĂĄnchez, P.M.S., Navarro-MartĂnez, D., LĂłpez, R.V., Lacal, J.I.R., Vivar, A.L., Monge, M.A.S., PĂ©rez, M.G., PĂ©rez, G.M.: Unlocking the potential of knowledge graphs: A cyber defense ontology for a knowledge representation and reasoning system. In: 19th International Conference on Availability, Reliability and Security (ARES). p. 73:1â73:9. ACM (2024). https://doi.org/10.1145/3664476.3670916 58. Valiente, M.C., MachĂn, R., GarcĂa-Barriocanal, E., Sicilia, M.Ă.: An Ontology-Based Inte- grated Approach to Situation Awareness for High-Level Information Fusion in C4ISR. In: Advanced Information Systems Engineering Workshops. p. 513â527. Springer (2011) 59. Wang, Z., Zhu, H., Liu, P., Sun, L.: Social engineering in cybersecurity: a domain ontology and knowledge graph application examples. Cybersecur. 4(1), 31 (2021). https://doi. org/10.1186/S42400-021-00094-6 60. Yongqin, H., Xushan, C., Anlian, Y., Shuo, P.: Research on application of knowledge graph in war archive based on big data. In: Big Data and Security. p. 234â247. Springer Nature (2023) 61. Zamazal, O., SvĂĄtek, V.: The ten-year ontofarm and its fertilization within the onto-sphere. J. Web Semant. 43, 46â53 (2017). https://doi.org/10.1016/J.WEBSEM.2017. 01.001