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 355 | Next

Robert Wrembel and Christian Koncilia

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

The value of each property
is computed as a bit string for each data instance, and these are concatenated to
form a pstring. To illustrate a PMap, consider an example relation Inventory with
attributes quantity, status, and product-category with a PMap having the following
properties:
??? Property.1:..Range on quantity; contains range intervals on the quantity attribute
from 20-29, 30-39, 40-49, and 49-59. Each interval is represented by a
bit string. For example, the interval 20-29 is represented by 00, 30-39 by 01,
40-49 by 10, and 50-59 by 11. Thus, Property 1 requires 2 bits.
??? Property.2: Enumerated on product-category; has the enumerated values of
the product-category attribute (1, 2, 3, 4 and 5). A bit string represents each
value. Therefore, 1 is represented 000, 2 by 001, 3 by 010, 4 by 011, and 5 by
100. Therefore, this property requires 3 bits.
??? Property.3:. Boolean on status = ???shipped,??? property value is true (or 1) when
a record??™s attribute status is ???shipped??? and false (or 0) otherwise. This property
requires 1 bit.
The pstrings for this example are 6 bits long; two bits for Property 1, three bits for
Property 2, and one bit for Property 3. Table 6 shows 5 records of an Inventory table
with the corresponding values for each property and the resulting pstring. The properties
are constructed based on some knowledge about the application (e.


Pages:
343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367