Article Journal of Information Science 2021, Vol. 47(2) 255–268 Ó The Author(s) 2019 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/0165551519887873 journals.sagepub.com/home/jis Karyon: A scalable and easy to integrate ontology summarisation framework Tugba Ozacar Department of Computer Engineering, Manisa Celal Bayar University, Turkey Ovunc Ozturk Department of Computer Engineering, Manisa Celal Bayar University, Turkey Abstract In the current Semantic Web Community, as the size and complexity of ontologies increase, ontology summarisation is becoming more important. There are many studies in the literature that use different approaches and metrics. However, many of these studies are not effective in terms of performance or have integration issues with current technologies. In this study, the popular ontology summarisation metrics are examined focusing on their performance in terms of time, and a number of metrics have been selected accordingly. To increase the accuracy of selections made with chosen metrics, we propose a novel metric: ‘name inclusion’. This metric promotes a concept if its name is subsumed by the name of another concept. As the existing summarisation applications have integra- tion issues, we have implemented our summarisation framework to integrate easily with the latest web technologies. Therefore, the algorithm is implemented using Rust language, which performs well and easily integrates with other languages. Keywords Key concepts; ontology; ontology summarisation; Rust 1. Introduction Ontology is a shared conceptual model of some domain of interest [1]. This model consists of concepts and relationships between these concepts. Ontologies have been used successfully in many scenarios in different application areas for the purpose of integrating knowledge bases. As a result, a large volume of linked semantic data was published. The Linked Open data cloud is an important project to illustrate the volume of published data. The dataset currently contains 1234 datasets with 16,136 links (as of June 2018). Such a large volume of data is difficult to understand and use by people. However, ontology understanding is quite necessary in terms of ontology engineering, ontology learning, ontology selec- tion, ontology matching and ontology reuse. In order to deal with such large volumes of data, it is a solution to use ontology summarisation approaches that filter out the most important parts of ontology. Ontology summarisation distills knowledge from an ontology and generates a bridge version to facilitate a better understanding. It is getting more attention recently [2,3]. Ontology summarisation algorithms are frequently used in ontology visualisation tools (KC-Viz [4], Onyx [5]), social network analysis [6,7] and semantic abstraction [8,9]. In this article, we propose a scalable algorithm to produce an ontology summary. Our approach uses a series of criteria drawn from network topology and lexical statistics. While choosing the criteria that willbe used in our algorithm, scal- ability was our main concern. We also proposed a new criterion to boost the accuracy of our algorithm. The proposed cri- terion, namely, ‘name inclusion’, is directly proportional to the amount of inclusion of the concept name in other concept names. Corresponding author: Ovunc Ozturk, Department of Computer Engineering, Manisa Celal Bayar University, Yunusemre, Manisa 45140, Turkey. Email: ovunc.ozturk@cbu.edu.tr