Tourism planning is becoming increasingly complex as travel behavior shifts from single-destination visits to multi-destination itineraries. However, many tourism information systems still rely on point-of-interest data and recommendation algorithms that lack the semantic structures needed to represent relationships between destinations. This limitation is important in smart tourism environments that require interoperable, data-integrated systems to support meaningful travel planning. This study conducts a systematic literature review of ontology and knowledge graph-based approaches for tourism planning to assess their ability to support itinerary modelling. Using PRISMA guidelines, relevant studies were retrieved from Scopus, ScienceDirect and IEEE Xplore, with 21 articles meeting all eligibility criteria. The review identifies six themes: ontology and semantic modelling, knowledge graph construction and enrichment, tourism recommendation and personalization, semantic retrieval and question answering, route planning and decision support, and itinerary-level and multi-destination representation. The analysis shows that ontology and knowledge graphs are widely used to organize tourism knowledge and enhance recommendation systems. However, most studies remain focused on entity-level modelling, while explicit semantic modelling of itinerary-level constructs remains limited. This gap highlights the need for ontology frameworks that can formally represent travel sequences and inter-destination dependencies to better support smart-tourism planning. The review identifies six themes: ontology and semantic model- ling, knowledge graph construction and enrichment, tourism recommendation and personalization, semantic retrieval and question answering, route planning and decision support, and itinerary-level and multi-destination representation. The analysis shows that ontology and knowledge graphs are widely used to organize tourism knowledge and enhance recommendation systems. However, most studies remain focused on entity-level modelling, while explicit semantic modelling of itinerary-level constructs remains limited. Therefore, studies on single-destina- tion, attraction-level, accommodation-level, cultural tourism, recom- mendation, or semantic retrieval systems were retained when they FIGURE 1 Conceptual framework integrating ontological foundation, multi-destination tourism concept, and systematic review methodology. Izani and Che Lah 10.3389/frai.2026.1881434 Frontiers in Artificial Intelligence 04 frontiersin.org used ontology or knowledge graph approaches and helped assess the extent of itinerary-level representation. The coding focused on the type of semantic approach, tourism application, modelling structure, evaluation method, and level of itinerary representation. The coding process involved three steps. First, each study was coded based on its main semantic approach, such as ontology engi- neering, knowledge graph construction, semantic retrieval, recom- mendation, or route planning. Second, studies were coded according to their tourism application, such as cultural heritage, accommoda- tion, attraction recommendation, hospitality, decision support, or itinerary-related planning. This pattern supports the review finding that tourism semantic sys- tems are commonly used for data organization, retrieval, and recom- mendation, but less often for formal representation of itinerary-level structures such as Itinerary and Route. 3.3.1 Ontology and knowledge graph-based approaches in tourism planning The thematic distribution in Table 5 reveals that ontology and knowledge graph-based approaches are used across different tourism planning and related tourism applications. The studies are not limited to multi-destination planning systems. They include ontology devel- opment, tourism knowledge graph construction, semantic retrieval, question answering, recommendation, route planning, and decision- support applications. This wider coverage is relevant because it shows how semantic technologies are currently used in tourism and whether they have progressed toward itinerary-level representation. At the ontology-centric end, studies such as Haridy et al. (2023) treat ontology construction as the principal outcome, emphasizing methodological transparency through ontology-driven conceptual modelling and NLP-supported term extraction. Pinto et al. Ontology-oriented studies tend to describe semantic structures more Izani and Che Lah 10.3389/frai.2026.1881434 Frontiers in Artificial Intelligence 09 frontiersin.org TABLE 4 Summary of ontology-based tourism studies and their characteristics. Study Author (year) Approach type Semantic method Application focus Evaluation type Itinerary-level representation S1 Zhang (2026) KG construction Knowledge graph, multimodal model Cultural tourism integration System/model evaluation Not explicit S2 Wang et al. (2024) KG-enhanced application KG, language models Tourism accommodation offers System evaluation Not explicit KG, knowledge graph; RDF , resource description framework; OWL, web ontology language; QA, question answering; GAT, graph attention network; GA, genetic algorithm; LLM, large language model. Itinerary-level representation was classified as explicit, partial, or not explicit. Explicit refers to direct modelling of itinerary-related constructs such as trip, itinerary, segment, temporal relation, route, sequence, or transition. (2024) ✓ Theme Description Number of studies Percentage T1 Ontology and semantic modelling 6 28.6% T2 Knowledge graph construction and enrichment 15 71.4% T3 Tourism recommendation and personalization 8 38.1% T4 Semantic retrieval and question answering 3 14.3% T5 Route planning and decision support 4 19.0% T6 Itinerary-level and multi-destination representation 3 14.3% A study may address multiple themes; therefore, percentages do not sum to 100%. ✓ indicates the study addresses the corresponding theme.

Knowledge Representation Issues in Semantic Graphs for Relationship Detection An important task for Homeland Security is the prediction of threat vulnerabilities, such as through the detection of relationships between seemingly disjoint entities. A structure used for this task is a "semantic graph", also known as a "relational data graph" or an "attributed relational graph". These graphs encode relationships as "typed" links between a pair of "typed" nodes. Indeed, semantic graphs are very similar to semantic networks used in AI. The node and link types are related through an ontology graph (also known as a schema). Furthermore, each node has a set of attributes associated with it (e.g., "age" may be an attribute of a node of type "person"). Unfortunately, the selection of types and attributes for both nodes and links depends on human expertise and is somewhat subjective and even arbitrary. This subjectiveness introduces biases into any algorithm that operates on semantic graphs. Here, we raise some knowledge representation issues for semantic graphs and provide some possible solutions using recently developed ideas in the field of complex networks. [cs/0504072] Knowledge Representation Issues in Semantic Graphs for Relationship Detection Knowledge Representation Issues in Semantic Graphs for Relationship Detection Marc Barthélemy 1 1 1 Authors listed alphabetically. CEA-Centre d’Etudes de Bruyères-Le-Châtel Departement de Physique Théorique et Appliquée BP12, 91680 Bruyères-Le-Châtel Cedex, France Edmond Chow Center for Applied Scientific Computing Lawrence Livermore National Laboratory Box 808, L-560, Livermore, CA 94551, USA Biodefense Knowledge Center, Lawrence Livermore National Laboratory. The node and link types are related through an ontology graph (also known as a schema ). Furthermore, each node has a set of attributes associated with it (e.g., “age” may be an attribute of a node of type “person”). Unfortunately, the selection of types and attributes for both nodes and links depends on human expertise and is somewhat subjective and even arbitrary. This subjectiveness introduces biases into any algorithm that operates on semantic graphs. Here, we raise some knowledge representation issues for semantic graphs and provide some possible solutions using recently developed ideas in the field of complex networks. This is a node type hierarchy that will be briefly mentioned when we discuss the scale of semantic graphs. Figure 1: A small ontology consisting of three node types. III Transitivity for Evaluating Nodes and Edges Consider a node “San Francisco” of type “city” in a semantic graph, and suppose we have a database of people which includes city of birth among the data fields. A node “Alice” of type “person” may be linked to the node “San Francisco” if Alice was born in San Francisco. Other nodes linked to node San Francisco imply a relationship to San Francisco and in turn their relation to Alice. Figure 2: A particular ontology for which neighbors of α 𝛼 \alpha of type δ 𝛿 \delta can never be connected to neighbors of type β 𝛽 \beta or γ 𝛾 \gamma . IV Statistical Measures for Semantic Graphs Along with clustering coefficient, two other relevant graph properties that have been developed for standard (non-semantic) graphs are distributions of node degree (number of neighbors of a node) and average path length between any two nodes in the graph. Together, these three graph properties can be useful for studying the properties of a semantic graph for representing knowledge. V Scale in Semantic Graphs Given a knowledge base of relational data, the choice of ontology depends on what information needs to be captured in the semantic graph, and how easily certain information needs to be retrieved. The level of detail (or scale) chosen for the ontology (choice of node and link types) will have a direct impact on the properties of the corresponding semantic graph. In the simplest ontology, we have nodes of only one type. In the example of the movies database, this ontology is a simple network of actors without any types and two actors are connected if they played in the same movie. At the next finer scale, we have actors and movies as node types. In the semantic graph, the nodes with the largest clustering coefficients depend on whether the types of the nodes are considered. In the standard case where the types are not considered, the node Maurice Barrymore has high clustering coefficient; the node is connected to Georgiana Drew Barrymore, Lionel Barrymore, Ethel Barrymore, etc., all of which are connected to each other. If node types are considered, then it is not This is consistent with nodes of types 1, 2, and 3 being of type “location,” nodes of type 28 being of type “terrorist organization,” and nodes of type 50 being of type “number.” The remaining types are types of attacks and are not particularly correlated with any other node types (given the numbers of each node type). We note in this case that semantically similar node types have similar values of m α subscript 𝑚 𝛼 m_{\alpha} and R ​ ( α ) 𝑅 𝛼 R(\alpha) . VII Conclusion This paper reveals some of the knowledge representation issues associated with semantic graphs. Ideas from the field of complex networks have been applied and generalized to semantic graphs. For example, transitivity may be used to determine the relevance of edge types for relationship detection. We have defined several measures for statistically characterizing node types. These quantities take into account the ontology which specifies the permitted connections in the semantic graph. Many other important measures can be defined, such as correlations with attribute values Jensen & Neville ( 2002 ) , which was not covered in this paper. These and other tools can be useful to help design ontologies and semantic graphs for knowledge representation. VIII Acknowledgments We are pleased to thank Keith Henderson and David Jensen for helpful discussions.

Ontology is an interdisciplinary field that involves both the use of philosophical principles and the development of computational artifacts. As artifacts, ontologies can have diverse applications in knowledge management, information retrieval, and information systems, to mention a few. They have been largely applied to organize information in complex fields like Biomedicine. In this article, we present the OntoNeo Ontology, an initiative to build a formal ontology in the obstetrics and neonatal domain. OntoNeo is a resource that has been designed to serve as a comprehensive infrastructure providing scientific research and healthcare professionals with access to relevant information. The goal of OntoNeo is twofold: (a) to organize specialized medical knowledge, and (b) to provide a potential consensual representation of the medical information found in electronic health records and medical information systems. To describe our initiative, we first provide background information about distinct theories underlying ontology, top‐level computational ontologies and their applications in Biomedicine. Then, we present the methodology employed in the development of OntoNeo and the results obtained to date. Finally, we discuss the applicability of OntoNeo by presenting a proof of concept that illustrates its potential usefulness in the realm of healthcare information systems.

Ontology-Based Users & Requests Clustering in Customer Service Management System Customer Service Management is one of major business activities to better serve company customers through the introduction of reliable processes and procedures. Today this kind of activities is implemented through e-services to directly involve customers into business processes. Traditionally Customer Service Management involves application of data mining techniques to discover usage patterns from the company knowledge memory. Hence grouping of customers/requests to clusters is one of major technique to improve the level of company customization. The goal of this paper is to present an efficient for implementation approach for clustering users and their requests. The approach uses ontology as knowledge representation model to improve the semantic interoperability between units of the company and customers. Some fragments of the approach tested in an industrial company are also presented in the paper. Published as: Smirnov A., Pashkin M., Chilov N., Levashova T., Krizhanovsky A., Kashevnik A. 2005. Ontology-Based Users and Requests Clustering in Customer Service Management System.

| 1 | Data and Knowledge Engineering Data Mining and Knowledge Discover Springer-Verlag … processes of knowledge gathering, knowledge representation, and knowledge use. In more general terms, the … and Jacob Weisberg .......::cce1008 221 ICT and Knowledge Management Systems / Irma Becerra-Fernandez

computer science and artificial intelligence are interested in concepts as formal structures involved in knowledge representation and automated reasoning A concept is a fundamental unit of cognition that classifies entities and encodes shared features. Concepts make it possible to form and combine ideas, draw inferences, and refer to external objects. They act as the meanings of words and play a central role in many cognitive processes, including perception, memory, and reasoning. Researchers distinguish different types of concepts based on their i Neurosc…

particles, lions, and stars. In the fields of computer science, information science, and knowledge representation, applied ontology is interested in the Ontology is the philosophical study of being. It is traditionally understood as the subdiscipline of metaphysics focused on the most general features of reality. As one of the most fundamental concepts, being encompasses all of reality and every entity within it. To articulate the basic structure of being, ontology examines the shared characteristics among all things and investigates their classif Ontologies are often used in information science to provide a conceptual scheme or inventory of a specific domain, making it possible to classify objects and formally represent information about them. This is of specific interest to computer science, which builds databases to store this information and defines computational processes to automatically transform and use it. For instance, to encode and store information about clients and employees in a database, an organization may use an ontology with categories such as person, company, address, and name. In some cases, it is necessary to exchange information belonging to different domains or to integrate databases using distinct ontologies. This can be achieved with the help of upper ontologies, which are not limited to one specific domain. They use general categories that apply to most or all domains, like Suggested Upper Merged Ontology and Basic Formal Ontology. Similar applications of ontology are found in various fields seeking to manage extensive information within a structured framework. Protein Ontology is a formal framework for the standardized representation of protein-related entities and their relationships. Gene Ontology and Sequence Ontology serve a similar purpose in the field of genetics. Environment Ontology is a knowledge representation focused on ecosystems and environmental processes. Friend of a Friend provides a conceptual framework to represent relations between people and their interests and activities. The topic of ontology has received increased attention in anthropology since the 1990s, sometimes termed the "ontological turn". This type of inquiry is focused on how people from different cultures experience and understand the nature of being. Specific interest has been given to the ontological outlook of Indigenous people and how it differs from a Western perspective. As an example of this contrast, it has been argued that various indigenous communities ascribe intentionality to non-human entities, like plants, forests, or rivers. This outlook is known as animism and is also found in Native American ontologies, which emphasize the interconnectedness of all living entities and the…