As an illustration, the following rules (CMAR and NCMAR respectively) were
extracted from the fact tables shown in Table 1 (left and right sides respectively)
by first computing FCIs.
??? CMAR:.Duality = 0 and Internal = 2 ?‡’ Govern = 2 [50/216, (50/216)/(74/216)].
This means that if the CEO does not act as a board chairman and if the proportion
of Top Management on the board is between 10 and 25%, then the index
264 Missaoui, Jatteau, Boujenoui, & Naouali
Copyright ?© 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of
Idea Group Inc. is prohibited.
of corporate governance quality is between 40 and 70% with a support = 23%
and a confidence = 67%.
??? NCMAR: Duality = 0 and Internal = 1 ?‡’ AvgAsset = 2 [73/216,
(73/216)/(74/216)]. This means that if the CEO does not act as a board chairman
and if the proportion of top management on the board is less than 10%,
then the average asset is over 1000K (according to the three defined dimensions)
with a support = 34% and a confidence = 98%.
Data.Warehousing.Techniques............
for.Data.Mining
Many researchers recognize the need to have mechanisms for user exploration and
guidance while mining from databases (Imielinski et al., 2002) or browsing through
data cubes (Sarawagi et al., 1998).
Since the output of a data mining task can be very large even for a reasonably small
dataset, our main objective here is to allow the user to explore an already computed
DM output (a set of groupings/concepts in our case) in a discovery-driven manner
similar to what is offered by OLAP techniques.
Pages:
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493