A structure- and content-based multimedia information retrieval system for XML documents

Because the number of XML documents is dramatically increasing, we need to develop a multimedia information retrieval system which can support both the retrieval based on document structure and the retrieval based on image content. In order to support the structure-based retrieval, we design keyword, structure, element, and attribute index structures by indexing XML documents based on the basic element unit and implement them by using the o2store storage system. For supporting the content-based retrieval, we design and implement high-dimensional index structures for both color and shape features based on X-tree. Finally, we do the performance evaluation of our multimedia information retrieval system in terms of system efficiency, such as retrieval time, insertion time, and storage overhead.