COGNITIVE ASPECTS OF SEMANTIC DESKTOP TO SUPPORT PERSONAL INFORMATION MANAGEMENT by Danish Nadeem

Better Knowledge Management enhances our efficiency of work and gives satisfaction. Personal Information Management (PIM) tools support augmented productivity of user in an organizational role. The research for developing standard metrics to capture, understand and model user requirements needs high attention, the thesis focuses on Cognitive considerations of Personal Information Management and evaluating the Semantic Desktop: Gnowsis , looking up for a standard “utility function” based on user’s PIM behaviour and derive standard evaluation methods. A number of cognitively adequate design methodologies have been explored and suggested for developing user-centred Semantic Desktop. The thesis critically views different Philosophical theories in order to realize their application in Semantic Desktop prototypes. The theory of Mental Models is explored and its characteristics are elicited as an analytic tool to understand peoples behaviour for organising their personal information. Mental Model has been revisited to understand how people represent their mental images, assumptions and real-world concepts. Moreover, its possible application is suggested for the Semantic Desktop. The current status of the Semantic Desktop evaluation for PIM support is still in its infancy, there are not many publications, which implies that the work in this direction demands high research attention to claim the usability of the Semantic Desktops. The thesis is an initiative to bring an attention on usability issues of the Semantic Desktop from Cognitive Science point of view. The research has been conducted within the Gnowsis prototype, being developed at DFKI. Several approaches are being taken to the date to study PIM. But study of PIM behaviour within the Semantic Desktop framework would be a novel approach. In this work we explore several evaluation criteria which are to be tested for the Semantic Desktop developments. The thesis concludes with suggestions on formalizing Mental Models and the evaluation scenarios of the Semantic Desktop prototypes.