g., go from
sector to sector category) or ignore a subset of attributes.
??? Drill-down: go down the attribute hierarchy of one or many attributes or add
a new attribute (e.g., introduce CEO??™s education).
??? Slice: limit the analysis to a given attribute/property (e.g., proportion of internal
shareholders on the board).
??? Dice:.focus on a subset of attributes and/or objects (e.g., features about board
structure for firms working on some given industry sectors).
Additional exploratory operations can be defined (e.g, drill-through, nest-unnest, split).
In the sequel, we first provide a formal definition of the four key OLAP operations
by borrowing the well-known notations used in relational algebra. Then, algorithms
for implementing the projection and assembly are proposed and discussed.
Definition of OLAP Operations
In the following, we redefine four commonly known OLAP operations when they
apply to a DM output that takes the form of a concept lattice.
Toward Integrating Data Warehousing with Data Mining Techniques 26
Copyright ?© 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
Rollup
Let P ??† A be the attributes of a given relational (or fact) table on which the rollup
will be conducted either by climbing up attribute hierarchies or by dropping attributes,
and let L=B(O, A, R) be a concept lattice.
Pages:
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497