Applying fuzzy logic to measure completeness of a conceptual model Tauqeer Hussain * , Mian M. Awais, Shafay Shamail Department of Computer Science, Lahore University of Management Sciences, DHA, Lahore 54792, Pakistan Abstract In a computing environment, the success of an information system depends upon the quality of its conceptual model. The importance of measuring quality of a conceptual model in quantitative terms has been emphasized in the research but still the quantitative measures are very scarce in the literature. A new Fuzzy Completeness Index (FCI) is introduced in this paper as a quantitative measure for the quality of a conceptual model. It takes into account completeness of a conceptual model based upon the concept of functional dependencies. For a given conceptual model the incorporation of functional dependencies is mapped onto a TAS Graph and is then measured using the fuzzy membership values and fuzzy hedges. The FCI has been calculated for different conceptual models. It has been illustrated that the quality in terms of completeness can effectively be measured through the FCI based approach. The higher the value of FCI the closer is the conceptual model to the real world in representing functional constraints. Ó 2006 Elsevier Inc. All rights reserved. Keywords: Fuzzy logic; Conceptual model; Quality; Completeness; Database design; Functional dependency 1. Introduction Traditional information systems are based on the precise data values that an attribute can take on. On the other hand, real world concepts very often have certain degree of imprecision or uncertainty which traditional informational systems cannot handle [1]. This fact has a negative impact on the quality of information and, in turn, on the quality of decision making produced through traditional information systems. The integral part of an information system is a programming environment and a database environment. For programming envi- ronment, the application of fuzzy logic has been studied in [2]; whereas the concept of fuzzy databases, its sig- nificance and application have been discussed in [3,4]. Furthermore, conceptual modeling being one of the most important steps of the database design methodology has also been benefited by the fuzzy approach [5–8]. However, quantitative treatment using fuzzy logic has not been extended to measure the quality aspects of a conceptual model. The main focus of our research in this paper is to apply fuzzy logic to compute quality of a conceptual model quantitatively. 0096-3003/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.amc.2006.07.053 * Corresponding author. E-mail addresses: tauqeer@lums.edu.pk (T. Hussain), awais@lums.edu.pk (M.M. Awais), sshamail@lums.edu.pk (S. Shamail). Applied Mathematics and Computation 185 (2007) 1078–1086 www.elsevier.com/locate/amc