Personalization Mechanisms for Content Indexing, Search, Retrieval and Presentation in a Multimedia Search Engine

Interest on multimedia content analysis, archival, retrieval and delivery is steadily increasing. This interest emanates from the digital media explosion that we are witnessing. Among others, the growth of multimedia content is compounding the need for more effective methods for searching such information. The automated web search engines that are currently used rely only on text descriptions and as a result provide matches of poor quality in case of multimedia content. On the other hand, search engines that utilize low-level descriptors to support content-based retrieval focus on algorithmic approaches and not on the provision of services oriented at efficiently covering the user's needs. The scope of this paper is to present a personalized web-based multimedia search engine implemented in the Java programming language. The proposed system combines the characteristics of the current search engines as well as new innovative features which guarantee a) reduced response time b) improved search results c) adaptation to each user's preferences d) easy handling of multimedia content e) interactive multimedia searching f) support of context-based user groups. The analytic description of the system's mechanisms and features renders clear how the proposed search engine adapts its functionality to each user's individuality.