Proposed metrics for data accessibility in the context of linked open data Mahdi Zahedi Nooghabi and Akram Fathian Dastgerdi Ferdowsi University of Mashhad, Mashhad, Iran Abstract Purpose – One of the most important categories in linked open data (LOD) quality models is “data accessibility.” The purpose of this paper is to propose some metrics and indicators for assessing data accessibility in LOD and the semantic web context. Design/methodology/approach – In this paper, at first the authors consider some data quality and LOD quality models to review proposed subcategories for data accessibility dimension in related texts. Then, based on goal question metric (GQM) approach, the authors specify the project goals, main issues and some questions. Finally, the authors propose some metrics for assessing the data accessibility in the context of the semantic web. Findings – Based on GQM approach, the authors determined three main issues for data accessibility, including data availability, data performance, and data security policy. Then the authors created four main questions related to these issues. As a conclusion, the authors proposed 27 metrics for measuring these questions. Originality/value – Nowadays, one of the main challenges regarding data quality is the lack of agreement on widespread quality metrics and practical instruments for evaluating quality. Accessibility is an important aspect of data quality. However, few researches have been done to provide metrics and indicators for assessing data accessibility in the context of the semantic web. So, in this research, the authors consider the data accessibility dimension and propose a comparatively comprehensive set of metrics. Keywords Linked open data, Accessibility, Data quality, Semantic web, Data accessibility metrics, GQM approach Paper type Research paper Data accessibility metrics in the context of linked open data (LOD) Data quality is a multi-dimensional concept (Pipino et al., 2002), for which various definitions have been provided. In general, the definitions of data (or information) quality take either an intrinsic or a contextual view of information. In the intrinsic view, information properties are largely defined independent of a specific user, task, or application. In the context-based view, information quality is primarily defined relative to the user, the task, and the application of the information. Moreover, the literature adds a representational dimension to the current views. The representational dimension addresses the extent to which information presentation effectively facilitates interpretation and understanding (Nelson et al., 2005). Overall, the mostly cited definition for data quality in literature is fitness for use (Wang and Strong, 1996; Strong et al., 1997; Wang, 1998; Tayi and Ballou, 1998; Veregin, 1999). This definition emphasizes the importance of user judgment for accepting or rejecting data based on the usability of the data provided. In general, the idea of data or information quality depends on the actual use of data, the design of the system, and the production processes involved in data generation (Wand and Wang, 1996). To better understand what we mean by data quality, we first need to define data. Data are the primary base of information that describes real world objects in a format Program Vol. 50 No. 2, 2016 pp. 184-194 © Emerald Group Publishing Limited 0033-0337 DOI 10.1108/PROG-01-2015-0007 Received 26 January 2015 Revised 29 October 2015 Accepted 7 December 2015 The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/0033-0337.htm 184 PROG 50,2