Naive Rules Do Not Consider Underlying Causality - Data Mining: Foundations and Practice - page 223

factors might lead to a crisp description of whether an effect will occur. How-
ever, in our commonsense world, it is unlikely that all possible factors can be
known. In commonsense, every day reasoning, we use approaches that do not
require complete knowledge.
People recognize that a complex collection of elements causes a particular
effect, even if the precise elements of the complex are unknown. They may
not know what events are in the complex; or, what constraints and laws the
complex is subject to. Sometimes, the details underlying an event are known
to a fine level of detail, sometimes not. Generally, people are more successful in
reasoning about a few large-grain sized events than many fine-grained events.
Perhaps, this can transfer over to computational models of causality.
A lack of complete, precise knowledge should not be discouraging. People
do things in the world by exploiting our commonsense perceptions of cause
and effect. When trying to precisely reason about causality, we need complete
knowledge of all of the relevant events and circumstances. In commonsense,
every day reasoning, we use approaches that do not require complete knowl-
edge. Often, approaches follow what is essentially a satisficing paradigm.
Instead of being concerned with all of the fined grained detail, a better
approach may be to incorporate granulation using rough sets and/or fuzzy
sets to soften the need for preciseness. And then accept impreciseness in the
description. Each complex can be considered to be a granule. Larger complexes
can be decomposed into smaller complexes. Thus, going from large-grained to
small-grained.
Regardless of causal recognition and representation methodologies, it is
important to decision making to understand when association rules have a
causal foundation. This avoids naıve decisions and increases the perceived
utility of rules with causal underpinnings.

References

1. Asher, H [1983] Causal Modeling, Sage, Newbury Park, California
2. Blalock, H [1964] Causal Inferences in Nonexperimental Research, W.W. Nor-
ton, New York
3. Berry, W [1984] Nonrecursive Causal Models, Sage, Newbury Park, California
4. Dawid, A [1999] “Who Needs Counterfactuals” in Causal Models and Intelligent
Data Management (ed) A. Gammerman Springer, Berlin Heidelberg New York
5. Freeman, W [1995] Societies Of Brains , Lawrence Erlbaum, 1995
6. Gammerman, A (ed) [1999] Causal Models and Intelligent Data Management,
Springer, Berlin Heidelberg New York
7. Goodrich, M, Stirling, W, Boer, E [2000] “Satisficing Revisited,” Minds and
Machines, v 10, 79-109
8. Glymour, C [2001] The Mind's Arrows, MIT Press (Bradford), London
9. Hausman,

D [1988] Causal Asymmetries, Cambridge University Press,

Cambridge, U.K.
10. Hilborn, R, Mangel, M [1997] The Ecological Detective: Confronting Models With
Data, Princeton University Press, Princeton, NJ