Chapter 5 Ontological Evaluation and Validation Samir Tartir, I. Budak Arpinar, and Amit P. Sheth 5.1 Introduction Building an ontology for a specific domain can start from scratch (Cristani and Cuel, 2005) or by modifying an existing ontology (Gómez-Pérez and Rojas-Amaya, 1999). In both cases, techniques for evaluating the characteristics and the validity of the ontology are necessary. Not only such techniques might be useful during the ontology engineering process (Paslaru et al., 2006), they can also be useful to an end-user who is looking for an ontology that is suitable for her application domain. The user can select the best ontology according to her application needs among several ontologies (Sabou et al., 2005). Ontology evaluation is an important task that is needed in many situations. For example, during the process of building of an ontology, ontology evaluation is important to guarantee that what is built meets the application requirement. Fernández et al. (1999) presents a life cycle for ontologies (Fig. 5.1). The life cycle is mainly based on Software Engineering processes. Their cycle includes three sets of activities: Management (that includes control and quality control), technical (that includes tasks for building an ontology), and support (that includes activities that are performed at the same time as the technical tasks). In this methodology, ontology evaluation was presented as an ongoing process throughout the ontology lifecy- cle in both the management and the support activities to illustrate its importance. Ontology evaluation is also important in cases where the ontology is automatically populated from different resources that might not be homogeneous, leading to dupli- cate instances, or instances that are clustered according to their sources in the same ontology, both of which may decrease the usefulness of the ontology. For example, the search for semantic associations (Anyanwu and Sheth, 2003) between entities in ontologies has been a major focus for the semantic web. These associations capture the complex relationships between entities that might be involve several other enti- ties and can’t be easily captured by human users in the midst of a large dataset. If a S. Tartir (B) Faculty of Information Technology, Philadelphia University, Amman 19392, Jordan e-mail: startir@philadelphia.edu.jo 115 R. Poli et al. (eds.), Theory and Applications of Ontology: Computer Applications, DOI 10.1007/978-90-481-8847-5_5, C  Springer Science+Business Media B.V. 2010