Database Reference
In-Depth Information
Figure 7. Geographic aggregation mode: Zones of a park
Example 3.
Let the aggregation of the geo-
graphic Measures S
zone
be represented by the
Entity
Schema
S
zone_agg
= ágeometry, nbDamagedTrees,
Fñ.
ϕ
1
(sum) to attribute a
1
, and the attribute b
2
is
the result of applying the operator ϕ
2
(union) to
attribute a
2.
GeoCube introduces the concept of
View
to
represent the result of a multidimensional query.
A
View
is a tuple of a
Base Cube
(the facts table),
the set of
Entity Schema
s used for the multidimen-
sional query, and an
Aggregation Mode
which
defines the
Entity
used as measure and aggregation
functions to aggregate it.
Example 4.
The multidimensional query
Where
and how many trees (which zones) have been
damaged by
some
fires during 1978?
” (Query 1,
cf.
Sec. “Research Motivations”) is represented
by the
View
V
zones-year
= áBC
naturalrisks
, áS
phenomenon
,
S
year
ñ, Θ
zone
, γñ. This query aggregates geographic
measures along the time dimenson. We note that
the
View
is composed of the
Base Cube
, the lev-
els used in the query S
phenomenon
(phenomena) and
S
ye
(years) and the
Aggregation Mode
Θ
zone
which
defines S
zone
as measure and establishes how to
aggregate it. The instance of V
zones-year
is shown
in Table 1.
The area is a metric attribute of the damaged
zone and so it can be directly and univocally
calculated using the geometry. For this reason,
it has to be considered as a derived measure.
Then, an example of
Geographic Aggregation
Mode
for the zones is Θ
zone
= áS
zone
, S
zone_agg
, Φ
union
ñ
where Φ
union
is the set of functions:
1. ϕ
geometry
: dom(S
zone
.geometry)
n
→
dom(S
zone_agg
.geometry), i.e. the Geometric
Union
2. ϕ
nbtrees
:dom(S
zone
.geometry, S
zone
.nbDam-
agedTrees)
n
→ dom(S
zone_agg
.nbDam-
agedTrees), i.e. the sum
An example of aggregation of two geographic
measures (Zone 3 and Zone 4) using Θ
zone
is
shown in figure 7. S
zone
is a schema entity with
attributes a
1
(the number of trees damaged by
fires), a
2
(the geometry of the region) and a a
3
area attribute a
3
(a
3
is derived from the geom-
etry). Zone 3 and Zone 4 in the figure are two
S
zone
entities. S
zone-aggreg
is another schema entity,
which represents the result of aggregating Zone
3 and Zone 4 (as shown in the dark grey area in
the right panel of the figure). The attribute b
1
in
S
zone-aggreg
is the result of applying the operator
correct geographic Aggregation
In order to formalize the constraints that ensure
the correct aggregation of geographic measures
according to the algorithm described in the Sec-
tion “Research Motivations”, we formalize the
semantics of alphanumeric attributes of geo-
Search WWH ::
Custom Search