Our paper “Semi-Supervised Online Structure Learning for Composite Event Recognition” was accepted at Machine Learning Journal
Evangelos Michelioudakis, Alexander Artikis and Georgios Paliouras
Online structure learning approaches, such as those stemming fromStatistical Relational Learning, enable the discovery of complex relations in noisydata 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 asa real dataset for maritime monitoring. The evaluation suggests that our approachcan effectively complete the missing labels and eventually, improve the accuracy ofthe underlying structure learning system.