Name
Abstract | ||
|---|---|---|
In this paper we propose a solution to the problem of distinguishing between causal and acausal temporal sets of rules. The method, called the Temporal Investigation Method for Enregistered Record Sequences (TIMERS), is explained and introduced formally. The input to TIMERS consists of a sequence of records, where each record is observed at regular intervals. Sets of rules are generated from the input data using different window sizes and directions of time. The set of rules may describe an instantaneous relationship, where the decision attribute depends on condition attributes seen at the same time instant. We investigate the temporal characteristics of the system by changing the direction of time when generating temporal rules to see whether a set of rules is causal or acausal. The results are used to declare a verdict as to the nature of the system: instantaneous, causal, or acausal. |
Year | DOI | Venue |
|---|---|---|
2003 | 10.1007/3-540-36175-8_23 | PAKDD |
Keywords | Field | DocType |
acausal temporal relation,temporal rule,acausal temporal set,decision attribute,input data,temporal investigation method,temporal characteristic,distinguishing causal,time instant,enregistered record sequences,instantaneous relationship,different window size | Artificial life,Data mining,Causality,Instant,Situation calculus,Computer science,Subject-matter expert,Artificial intelligence | Conference |
Volume | ISSN | ISBN |
2637 | 0302-9743 | 3-540-04760-3 |
Citations | PageRank | References |
11 | 1.07 | 5 |
Authors | ||
2 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Kamran Karimi | 1 | 118 | 17.23 |
| Howard J. Hamilton | 2 | 1501 | 145.55 |