INEX 2005 Multimedia Track

This paper reports on the activities of the INEX 2005 Multimedia track. The track was successful in realizing its objective to provide a pilot evaluation platform for the evaluation of retrieval strategies for XML-based multimedia documents. In this first exploratory year the focus of the evaluation experiment was to test approaches for the retrieval of XML fragments using a combination of content-based text and image retrieval techniques. The track is set to continue at INEX 2006. 1 Challenge and Objectives The main objective of the INEX 2005 multimedia track was to provide a pilot evaluation platform for structured document retrieval systems which are not limited to textual content, but combine multiple media types. Many, real-life structured document collections today contain a range of media such as text, image, speech, and video. Incorporating different media types within the retrieval process and producing meaningful rankings of multimedia documents and components is far from trivial. The goal of the multimedia track at INEX 2005 was to investigate this problem from a new perspective, using the structure of documents as the semantic/logical backbone for the retrieval of multimedia document fragments. In this first year, the track resulted in the construction of a pilot evaluation platform for the retrieval of multimedia structured document fragments. The methodology used in the construction of the test collection was built on established methods used at TREC [1]. The task set for the multimedia track was to retrieve relevant document fragments based on an information need with a structured multimedia character. The challenge for a structured document retrieval system in this case is then to combine the relevance of the different media types into a single (meaningful) ranking that can be presented to the user. The INEX multimedia track differs from other approaches in multimedia information retrieval, like TRECVID [2] and ImageCLEF [3], in the sense that it focuses on the use of document structure to estimate, relate, and combine the relevance of different multimedia fragments. The focus for 2005 was on the combination of text and image retrieval, where a strict interpretation of the structural constraints within the specified information need was adopted.