![]() | Only 14 pages are availabe for public view |
Abstract This thesis proposes the use of Knowledge Discovery in two domains, classification and content based image retrieval, medical databases is a very good application of classification; medical experts can express their concerns and preferences to guide knowledge exploration from the data sets. On applying the derived knowledge patterns in medical work, the domain experts can further justify the decision support information and then refine the scope of the knowledge. We proposed a generic classification framework that can be applied for manufacturing, business and medical applications.We used GA with fuzzy-gene individuals to implement variable length chromosomes, the output is a rule based classifier which is able to classify new cases. For content based image retrieval we proposed a new technique based on statistical moments and entropy to identify images using query-by-example queries. We used the developed technique for Arabic character recognition considering all possible ArabicThis thesis proposes the use of Knowledge Discovery in two domains, classification and content based image retrieval, medical databases is a very good application of classification; medical experts can express their concerns and preferences to guide knowledge exploration from the data sets. On applying the derived knowledge patterns in medical work, the domain experts can further justify the decision support information and then refine the scope of the knowledge. We proposed a generic classification framework that can be applied for manufacturing, business and medical applications.We used GA with fuzzy-gene individuals to implement variable length chromosomes, the output is a rule based classifier which is able to classify new cases. For content based image retrieval we proposed a new technique based on statistical moments and entropy to identify images using query-by-example queries. We used the developed technique for Arabic character recognition considering all possible Arabic character cases. |