Discovering Recurrent Events in Multichannel Data Streams Using Unsupervised Methods

Detection of recurrent temporal patterns in digital media is a first step in the next generation data mining for media content. Production videos such as news, sports and movies have a definitive structure that involves short term interaction as well as long term correlation. This structure in video can be captured by models that take into consideration the short term statistics as well as long term recurrence. We investigate the application of probabilistic models that capture this structure. The novel approach is to characterize the short term events in video by models that can account for temporal support in terms of piece-wise stationary signals with transitions. These short term events can then be embedded within another temporal model that accounts for transitions between these event and thus characterizes long term history. This also leads to the detection of recurring events in video using a monolithic model. The proposed approach is an unsupervised algorithm for event detection and it can be used for summarization, similarity based matching and enhanced browsing. With certain extensions similar algorithms can be used for data mining for temporal patterns in other domains such as bio-surveillance, where the multimodal sensor streams may be comprised of traditional and non-traditional data sources.