Semi-supervised online structure learning for composite event recognition

DOI10.1007/S10994-019-05794-2MaRDI QIDQ2425248zbMATH OpenOpenAlexFDO

Authors Evangelos Michelioudakis, Alexander Artikis, Georgios Paliouras

Publication date 26 June 2019

Published in Machine Learning (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1803.00546

zbMATH Keywords

semi-supervised learningevent recognitionevent calculusfirst-order logic distancegraph-cut minimisationonline structure learning

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Online algorithms; streaming algorithms (68W27)

Abstract: Online structure learning approaches, such as those stemming from Statistical Relational Learning, enable the discovery of complex relations in noisy data streams. However, these methods assume the existence of fully-labelled training data, which is unrealistic for most real-world applications. We present a novel approach for completing the supervision of a semi-supervised structure learning task. We incorporate graph-cut minimisation, a technique that derives labels for unlabelled data, based on their distance to their labelled counterparts. In order to adapt graph-cut minimisation to first order logic, we employ a suitable structural distance for measuring the distance between sets of logical atoms. The labelling process is achieved online (single-pass) by means of a caching mechanism and the Hoeffding bound, a statistical tool to approximate globally-optimal decisions from locally-optimal ones. We evaluate our approach on the task of composite event recognition by using a benchmark dataset for human activity recognition, as well as a real dataset for maritime monitoring. The evaluation suggests that our approach can effectively complete the missing labels and eventually, improve the accuracy of the underlying structure learning system.

Recommendations

• Online reliable semi-supervised learning on evolving data streams
• Online learning of event definitions
• Incremental learning of event definitions with inductive logic programming
• scientific article; zbMATH DE number 5957307
• Semi-supervised classification via simultaneous label and discriminant embedding estimation

Cites work

• scientific article; zbMATH DE number 1222289 (Why is no real title available?)
• scientific article; zbMATH DE number 1759248 (Why is no real title available?)
• scientific article; zbMATH DE number 3231692 (Why is no real title available?)
• scientific article; zbMATH DE number 2243385 (Why is no real title available?)
• scientific article; zbMATH DE number 3096290 (Why is no real title available?)
• A logic-based calculus of events
• Adaptive subgradient methods for online learning and stochastic optimization
• Clausal discovery
• Handbook of knowledge representation.
• Integrating naïve Bayes and FOIL
• Introduction to semi-supervised learning.
• Iterative cross-training: An algorithm for learning from unlabeled Web pages
• Logical and Relational Learning
• Markov logic networks
• Online learning of event definitions
• Parallel online learning of event definitions
• Probabilistic event calculus for event recognition
• Probability Inequalities for Sums of Bounded Random Variables
• Semi-supervised online structure learning for composite event recognition
• Top-down induction of first-order logical decision trees

Cited in

(5)

• A probabilistic interval-based event calculus for activity recognition
• Online event recognition over noisy data streams
• Semi-supervised online structure learning for composite event recognition
• Deep Fusion of Multiple Semantic Cues for Complex Event Recognition
• Making sense of sensory input

Describes a project that uses

Uses Software

• Aleph
• AdaGrad
• kFOIL

This page was built for publication: Semi-supervised online structure learning for composite event recognition

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2425248)