The world is witnessing an increasing use of IoT-based devices to collect sensor data in order to perceive the real world and generate abstractions. This data is highly heterogeneous in nature as it is obtained from various domains utilizing different representation schemes. Semantic approaches come as a rescue to this interoperability problem incurred because of heterogeneous sensor data from IoT devices. The data thus obtained should be represented in form of ontologies which are considered as the cornerstone of the Semantic Web for knowledge sharing, information extraction, information integration and many more. The content and the quality of the ontologies should be analyzed by evaluating them to ensure that the ontology is well designed, structured, and contains all essential concepts and relationships between them for efficient reasoning. This paper focuses on the evaluation of ontologies and, as a case study, evaluates a Military Resource Ontology (MRO) by using evaluation tools such as OntoMetric, OOPs!, ONTOCOM, based on evaluation approaches, aspects and criteria. These tools detect errors by diagnosing various metrics and pitfalls. Evaluation methods are grouped in two phases: verification and validation. In this paper, ‘QueryOnto’ tool is introduced to verify and validate the MRO by searching, query/answering, and visualizing.

Introduction

Because of significant attention from the research community, The Internet of Things (IoT) technology has been demonstrated to be a viable leading to an increased number of IoT applications. The primary goal of IoT is to interconnect devices and to collect data from them to create situational awareness and enable machines and humans better understand the situation or context hence facilitating better decision making and responses. IoT is being adapted in a large number of domains such as weather monitoring, water management, energy management, education, traffic control, retail, healthcare etc [1,2]. The IoT application space is quite heterogeneous and diverse due to which smart integration of data, applications and services has become a challenge in today’s world. The current approaches and technologies to IoT adaptation may not go a long way because of the evergrowing number of IoT devices and their diversity in representing data. Though web has a lot of information and is a convenient platform for integrating things, the machine cannot understand the intended meaning. Web 3.0 or simply semantic web further improves its capacity and capability of understanding the data coming from diverse IoT devices. The semantic web is the web of the future which helps machines understand the web page CONTACT Sanju Mishra sanju.tiwari.2007@gmail.com and makes life easier by facilitating the interoperability of various IoT devices. Semantic Web technologies are extensively employed to build interoperable IoT applications. Semantic approaches facilitate enhanced searches for data from voluminous and complex data sets from cloud [3]. All this is made possible because of the availability of common formats and metadata in the form of ontology. Ontologies provide a machine-understandable description of entities, relationships, and individuals. An ontology is a fundamental data structure for conceptualizing knowledge. Ontology can represent shared domain knowledge and enable semantic interoperability. It is used to capture the semantics of data in growing Semantic Web and treated as a backbone of semantic web applications and knowledge-based systems. Design and development of an ontology is a complex and ambiguous task, which takes a lot of effort and needs to be constantly reviewed to judge if concepts and relationships included are correct [4]. Generally, ontology keeps sufficient knowledge for achieving completeness to solve major problems, but evaluation is important to describe the ontology qualities and suitabilities for use. The Ontological model builds upon and extends the existing efforts in modeling and standardizing the IoT domain concepts and aims to capture most of the important relationships among those concepts. It is must to evaluate ontologies for ensuring the correctness and completeness of developing ontologies.