User Profiling Through Artificial Immune System (T-0672) (MFN 4226)

Abstract

An Enhanced Artificial Immune system(EAIS) is suggested in this thesis. The proposed system is an enhancement to the traditional Artificial Immune System (AIS) which will monitor user activities during search and will build searching profile for that user. We will use both implicit and explicit ways of declaring user profile and will maintain it according to five steps of AIS. The AIS will be modified by combining some extra parameters to make it more reliable and enable it to give nearest possible search results. This will enhance user experience while searching over Internet. It is also observed that some search engines provide misleading information about well known keywords and terms that cause invalid search results. To validate the performance of these new models simulation study is carried out by developing a search engine base supporting both AIS and EAIS. The results of simulation are compared with those obtained with Simple Searching algorithm and AIS algorithm. Improved identification and equalization performance of the proposed method have been observed in all cases.