State-sequence based pattern recognition in basketball zone-defense strategies

The notion of state is fundamental for many state-based computer systems, which represents the static snapshot of the world in discourse, while the dynamic historical scenarios of the world can be characterised in terms of temporally ordered state-sequences. This paper introduces a formal characterization of time-series and state-sequences, based on which, the formalism and algorithm for matching state-based temporal patterns are presented. As a case study of real-life applications, zone-defense pattern recognition in basketball games is specially examined as an illustrating example. Experimental results show that it is useful in helping the coach of the defense side to check whether the players play in a right zone-defense strategy, as well as the coach of the offensive side to detect the defense patterns of the opponent. Also, the approach proposed here can be adapted for applications to other team-work sports such as football, volleyball, and so on.