SEARCH
0-9 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Prev | Current Page 487 | Next

Robert Wrembel and Christian Koncilia

"Data Warehouses and Olap: Concepts, Architectures and Solutions"


Figure 3. Rollup of L1 by replacing c and d with a more general value k
268 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.
Drill-Down
In a symmetric way, Drill-down(L, P, AL) can be defined, where ljm in AL means
that the attribute hierarchy attached to Aj is to be explored from the current level
to a lower level ljm. Such operation uses L as input to produce a new concept lattice
resulting from the context in which the attributes in P are replaced with more
specific ones. When ljm = ^ for a given attribute Aj, this means that attribute Aj is a
new attribute to be considered for data mining.
The algebraic representation of Drill-down(L, P, AL) is the same as the one for the
rollup operation. The difference lies in the fact that for the drill-down operation, Al
is the set of P attributes specialized (through a move down in the attribute hierarchies)
according to the pattern AL.
Slice
Slice(L, ai) limits the analysis of a concept lattice L to a given attribute ai in order to
see its corresponding lattice. It is important to note that in data cubes, this operation
consists in selecting a given value for one dimension. However, we generalize it to
a given attribute (with its possible values) to allow the definition of new operations
(e.


Pages:
475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499