TY - GEN
T1 - Activity recognition with finite state machines
AU - Kerr, Wesley
AU - Tran, Anh
AU - Cohen, Paul
PY - 2011
Y1 - 2011
N2 - This paper shows how to learn general, Finite State Machine representations of activities that function as recognizers of previously unseen instances of activities. The central problem is to tell which differences between instances of activities are unimportant and may be safely ignored for the purpose of learning generalized representations of activities. We develop a novel way to find the "essential parts" of activities by a greedy kind of multiple sequence alignment, and a method to transform the resulting alignments into Finite State Machine that will accept novel instances of activities with high accuracy.
AB - This paper shows how to learn general, Finite State Machine representations of activities that function as recognizers of previously unseen instances of activities. The central problem is to tell which differences between instances of activities are unimportant and may be safely ignored for the purpose of learning generalized representations of activities. We develop a novel way to find the "essential parts" of activities by a greedy kind of multiple sequence alignment, and a method to transform the resulting alignments into Finite State Machine that will accept novel instances of activities with high accuracy.
UR - http://www.scopus.com/inward/record.url?scp=84881052291&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84881052291&partnerID=8YFLogxK
U2 - 10.5591/978-1-57735-516-8/IJCAI11-228
DO - 10.5591/978-1-57735-516-8/IJCAI11-228
M3 - Conference contribution
AN - SCOPUS:84881052291
SN - 9781577355120
T3 - IJCAI International Joint Conference on Artificial Intelligence
SP - 1348
EP - 1353
BT - IJCAI 2011 - 22nd International Joint Conference on Artificial Intelligence
T2 - 22nd International Joint Conference on Artificial Intelligence, IJCAI 2011
Y2 - 16 July 2011 through 22 July 2011
ER -