DFKI-LT - Mining over: Football Match Data: Seeking Associations among Explicit and Implicit Events

Jan Nemrava, Vojtech Svatek, Milan Simunek, Paul Buitelaar
Mining over: Football Match Data: Seeking Associations among Explicit and Implicit Events
2 Proc. of Znalosti 2007, Ostrava, Czech Republic, 2/2007
 
Football match data potentially cover a wide scope of resource modalities, such as video/audio streams, free-text news reports, structured online reports and structured data. One of the partial tasks in the K-Space project is to analyse multimedia resources together with resources that are complementary to them: texts and structured data. Eventually, the events (such as goals, fouls or substitutions) contained or automatically detected in different resources should be mapped to each other, so as to ease automated retrieval from 'difficult' sources via the 'easier' ones. In this paper we concentrate on the possibility of using association mining over each resource separately, however, with the aim of detecting relationships that could help an expert formulate criteria for future mapping across the resources.
 
Files: BibTeX, Znalosti.pdf