Towards Learning and Classifying Spatio-Temporal Activities in a Stream Processing Framework
2014 (English)In: STAIRS 2014: Proceedings of the 7th European Starting AI Researcher Symposium / [ed] Ulle Endriss and João Leite, IOS Press, 2014, 280-289 p.Conference paper (Refereed)
We propose an unsupervised stream processing framework that learns a Bayesian representation of observed spatio-temporal activities and their causal relations. The dynamics of the activities are modeled using sparse Gaussian processes and their causal relations using a causal Bayesian graph. This allows the model to be efficient through compactness and sparsity in the causal graph, and to provide probabilities at any level of abstraction for activities or chains of activities. Methods and ideas from a wide range of previous work are combined and interact to provide a uniform way to tackle a variety of common problems related to learning, classifying and predicting activities. We discuss how to use this framework to perform prediction of future activities and to generate events.
Place, publisher, year, edition, pages
IOS Press, 2014. 280-289 p.
, Frontiers in Artificial Intelligence and Applications, ISSN 0922-6389 (print) | 1879-8314 (online) ; 264
activity recognition, machine learning, knowledge representation, situation awareness, knowledge acquisition, unsupervised learning
IdentifiersURN: urn:nbn:se:liu:diva-110570DOI: 10.3233/978-1-61499-421-3-280ISI: 000350218400029ISBN: 978-1-61499-420-6 (print)ISBN: 978-1-61499-421-3 (online)OAI: oai:DiVA.org:liu-110570DiVA: diva2:746676
7th European Starting AI Researcher Symposium (STAIRS-2014), August 18-19, 2014, Prague, Czech Republic