New perspectives on the application of expert systems
M. Taboada, Rafael Martínez‐Tomás, José Manuel Ferrández Vicente · Expert Systems · 2011
Expert Systems (ES) are computer programs that use the knowledge and analytical skills (heuristics) of one or more human experts to infer solutions to problems in a particular discipline. The original aim of ES was to be able to replace human expertise (Buchanan, 1986). Nowadays, the exponential increase in the volume, complexity, and diversification of data has lead to the combination of different artificial intelligence techniques to support data access, analysis, and exploitation (Russell & Norvig, 2009). The representation and reasoning capabilities provided by rule-based ES are of assistance in tasks related to information access, interpretation, and management (Liao, 2005). Despite the relative maturity of such rule-based systems, they still contribute a great deal in diverse domains, such as medicine (Kong et al., 2009; Mabotuwana & Warren, 2009), telecommunication network design (Monedero et al., 2008), product configuration (Yang et al., 2009), and ontological engineering (Biletskiy & Girish, 2010). Moreover, the most recent applications have adopted a broad-minded approach to the subject, integrating expertise-based tasks with other Artificial Intelligence approaches, such as natural language processing (Demner-Fushman et al., 2009), case-based reasoning (Aamodt & Plaza, 1994) and event recognition (Fialho et al., 2010). The huge advances in computer technology have increased the possibilities for real applications using knowledge and heuristics, and has accelerated the growth of the efficient knowledge-based systems that are demanded in many practical domains, including as healthcare, surveillance, virtual environments, and system configuration. This special issue, entitled ‘New Perspectives on the Application of Expert Systems’, focuses on new applications using human expertise as an integral part for reasoning. It consists of extended versions of the best papers from the 3rd International Conference on the Interplay between Natural and Artificial Computation, (IWINAC 2009). The selected papers cover the use of human expertise in vital areas of modern ESs, and integrating this knowledge with other AI techniques, such as such as fuzzy reasoning, case-based reasoning, temporal reasoning, agent modelling, and visual system configuration. In the paper ‘Communication in distributed tracking systems: an ontology-based approach to improve cooperation’, Gómez-Romero et al. (2011) present a formal ontology aimed at the symbolic representation of visual data, specifically tracking information in a video-surveillance system. The ontology is used by a cooperative surveillance multi-agent system to increase the coordination and cooperation between independent and heterogeneous cameras. Additionally, the use of the ontology improves system scalability and facilitates the development of new functionalities. In the paper ‘Knowledge modelling through computational agents: Application to surveillance Systems’, Gascueña et al. (2011) model highly dynamic visual surveillance systems using computational agents. The novel underlying assumption in this work is that an agent starts being a conceptual model, then it is reduced to a formal model, and finally to a physical machine with sensors, effectors, and a control program. This assumption emphasizes the computable aspects of agent theory, allowing a higher degree of autonomy and response of agents because of their capabilities to adapt and to cooperate In the paper ‘T-CARE: Temporal Case Retrieval System’, Juarez et al. (2011) assume that the temporal evolution of the patient is a key factor in providing effective healthcare. This paper introduces T-CARE, an innovative temporal case retrieval system in the specific domain of Intensive Care Burns Unit. The system combines classical and non-classical approaches to measure temporal similarity of cases, which are composed of temporal sequences of time point events and intervals. In ‘Adaptive Fuzzy Knowledge-Based Multi-Agent Systems in Virtual Environments’, Arroyo et al. (2011) deal with pioneering aspects such as the confluence of 3D virtual worlds with social networks. The authors explore the possibility of using metabots, metaverse robots, in complex virtual 3D worlds, with motion capabilities based on an Adaptive Fuzzy Knowledge-Based controller, and driven by social issues. In the last paper ‘ARDIS: Knowledge-based architecture for visual system configuration in dynamic surface inspection’, Martin Gomez et al. (2011) present an original approach to dynamic surface inspection in laminated materials. The work is based on the configuration of a visual system in order to obtain good quality control of the manufacturing surface. It also aims to overcome some of the limitations of the single-use visual inspections systems, by integrating and differentiating knowledge. All these works represent the best contributions in ESs to the International World-conference on the Interplay between Natural and Artificial Computation (IWINAC-2009). We hope that the contributions of this special issue facilitate the interplay of proposals between Natural Sciences and Computation. Finally, we dedicate this special issue to the memory of Professor Mira. We would like to thank Dr. Jon G. Hall, the Editor-in-Chief of Expert Systems, for his interest and ongoing help for this special issue. This special issue, done in honour of José Mira, would not have been possible without the support of the Ministerio de Ciencia e Innovación through the projects TIN2007-67586-C02-01, TIN2009-14159-C05-05, and TIN2010-20845-C03-02. M. Taboada M. Taboada is Associate Professor of Computer Science and Artificial Intelligence at University of Santiago de Compostela. Her current research interests include knowledge engineering, ontology and terminology mapping, and archetype modeling in medicine. R. Martínez-Tomás R. Martínez-Tomás is Associate Professor of Computer Science and Artificial Intelligence at Spanish National University of Distance Learning (UNED). He obtained his PhD degree in Artificial Intelligence from UNED in 2000. He has worked on several projects related to artificial intelligence in medicine and video-sequence identification in surveillance tasks. His current research interests include knowledge engineering, knowledge based systems, spatial-temporal logics, description logics and video-sequence semantic interpretation. He is voluntarily serving as a technical publication reviewer for several respected scientific journals and conferences. J. M. Ferrández J. M. Ferrández is Associate Professor of Computer Science at Universidad Politécnica de Cartagena. He is the coordinator of Spanish Thematic Network RTNAC (rtnac.org) and the Iberoamerican network CANS, related to Natural and Artificial Computation. He is also the General Chairman of the International Conference IWINAC, International Work Conference on the Interplay between Natural and Artificial Computation.