Unsupervised pattern mining from symbolic temporal data
ACM SIGKDD Explorations Newsletter - Special issue on data mining for health informatics
Review: learning like a baby: A survey of artificial intelligence approaches
The Knowledge Engineering Review
Activity recognition with finite state machines
IJCAI'11 Proceedings of the Twenty-Second international joint conference on Artificial Intelligence - Volume Volume Two
| Hi-index | 0.00 |
Agents need to know the effects of their actions.Strongassociations between actions and effects can be found bycounting how often they co-occur.We present an algorithmthat learns temporal patterns expressed as fluents, propositionswith temporal extent.The fluent-learning algorithmis hierarchical and unsupervised. It works by maintainingco-occurrence statistics on pairs of fluents.In experimentson a mobile robot, the fluent-learning algorithm found temporalassociations that correspond to effects of the robot'sactions.