Abstract | ||
---|---|---|
In this paper, we revisit the problem of in- ducing a process model from time-series data. We illustrate this task with a realistic ecosys- tem model, review an initial method for its induction, then identify three challenges that require extension of this method. These in- clude dealing with unobservable variables, finding numeric conditions on processes, and preventing the creation of models that over- fit the training data. We describe responses to these challenges and present experimental evidence that they have the desired effects. After this, we show that this extended ap- proach to inductive process modeling can ex- plain and predict time-series data from bat- teries on the International Space Station. In closing, we discuss related work and consider directions for future research. |
Year | Venue | Keywords |
---|---|---|
2003 | ICML | process model,international space station,time series data |
Field | DocType | Citations |
Training set,Data mining,Time series,Computer science,Process modeling,Artificial intelligence,Overfitting,International Space Station,Unobservable,Machine learning,Ecosystem model | Conference | 11 |
PageRank | References | Authors |
0.71 | 13 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Pat Langley | 1 | 6471 | 1307.64 |
Dileep George | 2 | 154 | 12.69 |
Stephen D. Bay | 3 | 815 | 46.71 |
Kazumi Saito | 4 | 11 | 1.72 |