Integrating Large Language Models with Semantic Knowledge Graphs for Domain-Specific Conversational Systems: A Case Study in Tourism
Journal of Intelligent Decision Making and Information Science · 04.09.2026 · 3 (10s) · 2313-2327
DOI: 10.17762/jidmis.v3.2966 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Intelligent Decision Making and Information Science, online erschienen am 04.09.2026. Verfasst von Mohamad Elouzeir. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Chatbots are widely used for generating information for domain-specific applications like the tourism domain. However, the accuracy of the responses obtained is usually low or may be incomplete, especially if the chatbots are based only on general language models or unstructured data sources. In this paper, the use of Semantic Web technologies for improving the accuracy of the responses obtained from chatbots for answering queries related to the domain of tourism is discussed. A domain ontology and knowledge graph were designed based on the hotel dataset, enabling the use of SPARQL queries for accessing the information related to the domain of tourism. Two chatbot architectures were implemented for the domain of tourism. The first chatbot architecture is based on the GPT-based chatbot, while the second chatbot architecture is based on the rule-based chatbot using the Rasa framework. The evaluation criteria for the response were the response execution time, correctness, and the practical aspects of the system’s maintenance. As indicated by the results, the semantic chatbot based on the GPT model is more consistent and efficient, especially for the responses containing more attributes. Moreover, the comparison of the responses obtained by the GPT model with access to the knowledge graph and the responses obtained without access to the knowledge graph indicates the high potential of grounding the language models on the knowledge graph for obtaining more accurate responses.
Zitieren
APA
Elouzeir, M. (2026). Integrating Large Language Models with Semantic Knowledge Graphs for Domain-Specific Conversational Systems: A Case Study in Tourism. Journal of Intelligent Decision Making and Information Science, 3(10s), 2313-2327. https://doi.org/10.17762/jidmis.v3.2966
Vancouver
Elouzeir M. Integrating Large Language Models with Semantic Knowledge Graphs for Domain-Specific Conversational Systems: A Case Study in Tourism. Journal of Intelligent Decision Making and Information Science. 2026;3(10s):2313-2327. doi:10.17762/jidmis.v3.2966
BibTeX
@article{Elouzeir2026,
author = {Elouzeir, Mohamad},
title = {Integrating Large Language Models with Semantic Knowledge Graphs for Domain-Specific Conversational Systems: A Case Study in Tourism},
journal = {Journal of Intelligent Decision Making and Information Science},
year = {2026},
volume = {3},
number = {10s},
pages = {2313--2327},
doi = {10.17762/jidmis.v3.2966}
}
EUVIMED