Unsupervised Activity Discovery and Characterization From Event-Streams
This addresses the problem of automated activity recognition for applications in smart environments, but appears incremental as it builds on existing graph and Markov process methods.
The authors tackled the problem of discovering and characterizing everyday activity classes from event-streams in an unsupervised manner, achieving results that demonstrate the framework's competence and generalizability across multiple datasets.
We present a framework to discover and characterize different classes of everyday activities from event-streams. We begin by representing activities as bags of event n-grams. This allows us to analyze the global structural information of activities, using their local event statistics. We demonstrate how maximal cliques in an undirected edge-weighted graph of activities, can be used for activity-class discovery in an unsupervised manner. We show how modeling an activity as a variable length Markov process, can be used to discover recurrent event-motifs to characterize the discovered activity-classes. We present results over extensive data-sets, collected from multiple active environments, to show the competence and generalizability of our proposed framework.