Intelligent systems today are being applied in a range of areas including image processing, engineering, process control, data mining, web intelligence/web personalization, recommender systems, e-commerce, e-business, emotional intelligence and others. In the process of applying intelligent systems to complex real world problems several phenomena have emerged. Firstly, the application of intelligent agents in distributed environments has resulted in merger of techniques from soft computing area with those in distributed artificial intelligence. Secondly, the internet era in the last decade, among other aspects, has resulted in information explosion, and consequently the need for personalization of information has assumed significance. Thirdly, advancements made in multimedia and mobile technologies are leading to a need for context-aware systems. For example, besides factoring in location of a user, a need for integrating non-verbal behavior or inputs with other textual inputs of the user for developing smart systems has arisen. The need for context-awareness can also be seen in terms of need for intelligent systems to continuously adapt in uncertain and dynamic environments. Fourthly, from intelligent or soft computing technology viewpoint there is a need to improve the quality of solutions provided by soft computing paradigms in order to facilitate their acceptability in the industry. Fifthly, and importantly, from a practitioners or user's perspective, as these technologies have moved out of laboratories into the real world the need for practitioner or human-centered as against technology-centered approach to the design of intelligent systems has emerged. Finally, from a software engineering perspective, issues like scalability and maintainability need to be addressed before intelligent systems can be successfully deployed in the industry. In this seminar , I will outline a six-layered distributed multi-agent architecture which models the above phenomena. Thus the focus is on the methodology for the design of intelligent systems in complex data intensive domains. The multi-agent architecture is one of the four components of a human-centered intelligent system development framework. I will illustrate the application of the architecture with the help of examples in biomedicine, emotional intelligence, data mining, e-recommender systems, sensor networks and (time permitting) in other areas.
Last modified: Thursday, 28-Jul-2005 17:23:30 NZST
This page is maintained by the seminar list administrator.