g., Nassis, Rajugan, Dillon, & Rahayu, 2005; Park, Han, & Song, 2005).
Advanced image processing technologies make the use of images and maps easier
and more common, for example, Google Maps and NASA Earth Observing System
Data and Information System (EOSDIS). In order to use information hidden
in this kind of data a user needs a technology that combines the functionality of
Geographical Information Systems with the functionality of data warehouses and
OLAP. To this end, the so called Spatial OLAP systems are being developed (e.g.,
Stefanovic, Han, & Koperski, 2000).
Last but not least, our environment is becoming gradually filled with different kinds
of sensors, for example, monitoring the intensity of traffic, controlling physical
parameters of technological processes, and monitoring patients??™ vital signs. Sensors
produce streams of data that have to be analyzed, often online. Stream data arrive
also from other sources, for example, a click-stream form on the Internet, shares
from a stock market, and transmission signals in telecommunications. Stream data
are characterized by a continuous flow, requiring huge storage space. In order to
reduce the space, historical data are stored as summaries or samples. Typically,
incoming stream data need to be continuously analyzed. This leads to challenges in
(1) querying streams online, (2) querying both historical summarized/sampled data
and just incoming data, as well as (3) quickly accessing data, that is, indexing.
Pages:
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30