VideoCLEF is a new track for CLEF 2008. This track aims to develop and evaluate tasks in processing video content in a multilingual environment.

The VideoCLEF track for 2008 is centered on the Vid2RSS2008 task. This comprises a number of subtasks including topic classification performed on dual language videos. The longer term overall objective of VideoCLEF is to combine and extend these subtasks and perform completely automatic generation of RSS feeds specific to a particular information need and personalized to a particular language preference. The video data for Vid2RSS2008 are Dutch television documentaries and contain Dutch as a dominant language, but also contain a high proportion of spoken English (i.e., interview guests often speak in English).

The main task for Vid2RSS2008 involves assigning topic class labels to videos of television episodes. Speech recognition transcripts, metadata records (containing title and description) and video keyframes (and shot boundaries) for each episode are supplied. The output format is a set of RSS-feeds, one for each topic class. The RSS-feeds are created by concatenating the metadata records for the episodes assigned to a given topic class. (The RSS-feed format is trivial to generate and is used as an output format because it can be easily visualized in a feed aggregator.)

There is one mandatory task: classifying videos of television episodes using speech transcripts only.

There are several optional tasks: classifying episodes using both speech-transcripts and metadata (or potentially including visual content), translation of content to generate feeds in English (or another language of your choice), selection of keyframes to represent individual episodes, and shot boundary and keframe extraction from the video source.

The development data (10 episodes / ca. 6 hours) and sample classes will be released at the end of March along with sample topic categories. The test data (40 episodes /ca. 24 hours) will be released at end of April with the test topic categories. The submission date for the task will be set towards the end of June. The CLEF 2008 Workshop will be held on 17-19 September 2008 in Aarhus, Denmark, in conjunction with the 12th European Conference on Digital Libraries (ECDL 2008). More information about Vid2RSS2008 is available at: http://ilps.science.uva.nl/Vid2RSS/. Participants in VideoCLEF are required to register with CLEF. Registration for CLEF 2008 is now open, full information about CLEF can be found at: http://www.clef-campaign.org/.

Listening to ''For Energy Infinite'', by Mazarin (Play Count: 17)

Looking at Things Differently: Exploring Perspective Recall for Informal Text Retrieval by Wouter Weerkamp and Maarten de Rijke is available now. The paper will be presented at DIR 2008 this April; it reports on ongoing work where we examine the use of query expansion against multiple external corpora so as to uncover multiple perspective on a given topic. Our working assumption is that uncovering multiple perspectives is especially helpful when searching informal text (blogs, discussion forums, comments, etc).

Listening to ''Thin Blue Flame'', by Josh Ritter (Play Count: 0)

The proceedings versions of two CLEF 2007 papers are online now: Overview of WebCLEF 2007, by Valentin Jijkoun and Maarten de Rijke, and Using Centrality to Rank Web Snippets by the same authors. Also available now is Personal Name Resolution of Web People Search by Leif Azzopardi, Krisztian Balog and Maarten de Rijke; this paper will appear in the WWW 2008 workshop on NLP Challenges in the Information Explosion Era (NLPIX 2008).

Listening to ''Fate (Aka For Soph)'', by The Durutti Column (Play Count: 31)