Paper Info
Title
Discovering Temporal Rules from Temporally Ordered Data
Abstract
We introduce a method for finding temporal and atemporal relations in nominal, causal data. This method searches for relations among variables that characterize the behavior of a single system. Data are gathered from variables of the system, and used to discover relations among the variables. In general, such rules could be causal or acausal. We formally characterize the problem and introduce RFCT, a hybrid tool based on the C4.5 classification software. By performing appropriate preprocessing and postprocessing, RFCT extends C4.5's domain of applicability to the unsupervised discovery of temporal relations among temporally ordered nominal data.
Year
DOI
Venue
2002
10.1007/3-540-45675-9_5
IDEAL
Keywords
Field
DocType
hybrid tool,unsupervised discovery,method search,classification software,appropriate preprocessing,nominal data,discovering temporal rules,temporal relation,temporally ordered data,causal data,atemporal relation,single system
Decision rule,Data mining,Causality,Search algorithm,Computer science,Preprocessor,Bayesian network,Software,Unsupervised learning,Artificial intelligence,Machine learning
Conference
Volume
ISSN
ISBN
2412
0302-9743
3-540-44025-9
Citations 
PageRank 
References 
4
0.68
9
Authors
2
Name
Order
Citations
PageRank
Kamran Karimi111817.23
Howard J. Hamilton21501145.55