ABSTRACT
Part IV, consisting of five chapters, Chapters 12, 13, 14, 15, and 16, described the image classification tool that we developed. Chapter 12 provided an overview of approaches to managing and mining multimedia data. First, we discussed issues pertaining to handling multimedia data, and then we focused on initial data types such as text, images, audio, and video. The idea is to extract concepts from unstructured data and then mine them. Chapter 13 described various image classification models we have utilized in developing our image mining tool. In Chapter 14, we described a tool that we have developed which annotates the images. We also described the experimental results we have obtained and compared them with other approaches. In Chapter 15, we addressed the image annotation problem using enhanced feature weighting algorithms. We proposed aggressive and chi-square-based feature weighting mechanisms. In Chapter 16, we studied different KNN classifiers and modified EKNN algorithm and compared the performance results of these algorithms.
