UNIZULU Institutional Repository

User preference mining for context-aware M-services applications

Show simple item record

dc.contributor.advisor Adigun, M.O.
dc.contributor.advisor Xulu, S.S.
dc.contributor.author Jembere, Edgar
dc.date.accessioned 2010-07-22T09:07:45Z
dc.date.available 2010-07-22T09:07:45Z
dc.date.issued 2007
dc.identifier.uri http://hdl.handle.net/10530/354
dc.description A dissertation submitted in fulfillment of the requirements for the Degree of Master of Science in the Department of Computer Science, Faculty of Scince and Agriculture at the University of Zululand, 2007. en_US
dc.description.abstract Challenges to the field of Human Computer Interaction (HCI) arising from the emergence of mobile computing can be solved by tailoring the access and use of the mobile services to user preferences. User preferences are traditionally assumed to be static, but due to the dynamic nature of the mobile computing environment, this assumption no longer holds. In an m-Services environment user preferences are not only transient, but they also vary with the changes in context. Furthermore, the assumed preference models do not give an intuitive interpretation of a preference and lack user expressiveness. To address these issues, this research work defines a user preference model for a context-aware m-services environment, based on an intuitive quantitative preference measure and a strict partial order preference representation. We present some user preference mining algorithms and a framework for context-based user preferences mining in an m-Services environment. The developed user preference modelling and mining framework was prototyped and evaluated it terms of its quality and effectiveness. The user session data for the evaluation of the framework was generated using MATLAB 7.1's Generalised Pareto Probability Density Function (gppdf) with shape, scale and threshold parameters of 1.25,1, and 0 respectively. The framework was found to be relatively promising in terms of its effectiveness. The user preference mining framework was also found to relatively scale well with increases in the volumes of data. en_US
dc.language.iso en en_US
dc.subject Human computer interaction en_US
dc.subject Mobile computing en_US
dc.title User preference mining for context-aware M-services applications en_US
dc.type Thesis en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search UZSpace


Browse

My Account