Online learning of event definitions. Issue 5 (14th October 2016)

• Record Type: Journal Article
• Authors: KATZOURIS, NIKOS; ARTIKIS, ALEXANDER; PALIOURAS, GEORGIOS
• Editors: Carro, Manuel; King, Andy
• Is Part Of: Theory and practice of logic programming. Volume 16:Issue 5/6(2016)
• Page Start: 817
• Page End: 833
• Publication Date: 2016-10-14
• Subjects: Inductive Logic Programming, -- Event Calculus, -- Online Learning; Logic programming -- Periodicals; Artificial intelligence -- Computer programs -- Periodicals; Constraint programming (Computer science) -- Periodicals; 005.115
• DOI: 10.1017/S1471068416000260
• Languages: English
• ISSNs: 1471-0684

Abstract

Abstract: Systems for symbolic event recognition infer occurrences of events in time using a set of event definitions in the form of first-order rules. The Event Calculus is a temporal logic that has been used as a basis in event recognition applications, providing among others, direct connections to machine learning, via Inductive Logic Programming (ILP). We present an ILP system for online learning of Event Calculus theories. To allow for a single-pass learning strategy, we use the Hoeffding bound for evaluating clauses on a subset of the input stream. We employ a decoupling scheme of the Event Calculus axioms during the learning process, that allows to learn each clause in isolation. Moreover, we use abductive-inductive logic programming techniques to handle unobserved target predicates. We evaluate our approach on an activity recognition application and compare it to a number of batch learning techniques. We obtain results of comparable predicative accuracy with significant speed-ups in training time. We also outperform hand-crafted rules and match the performance of a sound incremental learner that can only operate on noise-free datasets.