Published in Geographic Data Mining and Knowledge Discovery,

Research Monographs in GIS, Taylor and Francis, 2001.

Algorithms and Applications for Spatial Data Mining

Martin Ester, Hans-Peter Kriegel, Jrg Sander (University of Munich)

1 Introduction
Due to the computerization and the advances in scientific data collection we are faced with a
large and continuously growing amount of data which makes it impossible to interpret all this data
manually. Therefore, the development of new techniques and tools that support the human in transforming data into useful knowledge has been the focus of the relatively new and interdisciplinary
research area knowledge discovery in databases.
Knowledge discovery in databases (KDD) has been defined as the non-trivial process of discovering valid, novel, potentially useful and ultimately understandable patterns from data, a pattern is
an expression in some language describing a subset of the data or a model applicable to that subset
(Fayyad et al., 1996). The process of KDD is interactive and iterative, involving several steps such
as data selection, data reduction, data mining, and the evaluation of the data mining results. The
heart of the process, however, is the data mining step which consists of the application of data analysis and discovery algorithms that, under acceptable computational efficiency limitations, produce
a particular enumeration of patterns over the data (Fayyad et al., 1996).
While a lot of research has been conducted on knowledge discovery and data mining in relational databases (see e.g. (Chen et al., 1996) or (Fayyad, 1997) for an overview), only a few works
deal with knowledge discovery in spatial databases (see (Gueting, 1994) for an introduction to spatial databases, (Koperski et al., 1996) for an overview of spatial data mining). Finding implicit regularities, rules or patterns hidden in spatial databases is an important task, e.g. for geo-marketing,
traffic control or environmental studies.
A spatial database contains objects which are characterized by a spatial location and/or extension as well as by several non-spatial attributes. Figure 7.1 illustrates a spatial database on Bavaria
as an example. Depicted is the relation Communities containing polygons which represent communities in a geographic information system. This spatial database on Bavaria - referred to as the BAVARIA database - is used in some of the following sections as test database for our algorithms. The

database contains the ATKIS 500 data and the Bavarian part of the statistical data obtained by the
-1-

Relation Bavarian Communities

name

population

unemployment

foreigners

...

Munich

1.300.000

0.06

0.15

...

...

...

...

...

...

spatial

...

Figure 7.1. Spatial and non-spatial attributes of communities

German census of 1987, i.e. 2043 Bavarian communities with one spatial attribute (polygon) and
52 non-spatial attributes (such as average rent or rate of unemployment). Also included (in a separate table of the database) are spatial objects representing natural objects like mountains or rivers
and infrastructure such as highways or railroads. The total number of spatial objects in the database
then amounts to 6924.
The discovery process for spatial data is more complex than for relational data. This applies to
both the efficiency of algorithms as well to the complexity of possible patterns that can be found
in a spatial database. The reason is that, in contrast to mining in relational databases, spatial data
mining algorithms have to consider the neighbours of objects in order to extract useful knowledge.
This is necessary because the attributes of the neighbours of some object of interest may have a
significant influence on the object itself.
In this chapter, we will first introduce, in section 7.2, a general framework for spatial data mining which takes into account the mentioned characteristics of spatial data. We also show that this
approach allows a tight and efficient integration of spatial data mining algorithms with spatial database systems. In section 3 through 6, we then present algorithms for the tasks of spatial clustering, spatial characterization, spatial trend detection and spatial classification utilizing the proposed

-2-

framework. Furthermore, example applications are discussed for these algorithms. The last section
gives a short summary and shows some directions for future research.

2 A Database-Oriented Framework for Spatial Data Mining

Our framework for spatial data mining is based on spatial neighbourhood relations between objects and on the induced neighbourhood graphs and neighbourhood paths which can be defined
with respect to these neighbourhood relations. Thus, we introduce a set of database primitives or
basic operations for spatial data mining which are sufficient to express most of the spatial data mining algorithms from the literature. This approach has several advantages. Similar to the relational
standard language SQL, the use of standard primitives will speed-up the development of new data
mining algorithms and will also make them more portable. Second, we can develop techniques to
efficiently support the proposed database primitives (e.g. by specialized index structures) thus
speeding-up all data mining algorithms which are based on our database primitives. Moreover, our
basic operations for spatial data mining can be integrated into commercial database management
systems. This will offer additional benefits for data mining applications such as efficient storage
management, prevention of inconsistencies, index structures to support different types of database
queries which may be part of the data mining algorithms.
2.1 Spatial Neighbourhood Relations, Spatial Neighbourhood Graphs and their Operations
Our database primitives for spatial data mining are based on the concepts of neighbourhood
graphs and neighbourhood paths which in turn are defined with respect to neighbourhood relations
between objects (Ester et al., 1997).
There are three basic types of spatial relations: topological, distance and direction relations
which may be combined by logical operators to express a more complex neighbourhood relation.
Spatial objects such as points, lines, polygons or polyhedrons are all represented by a set of points.
For example, a polygon can be represented by its edges (vector representation) or by the points contained in in its interior, e.g. the pixels of an object in a raster image (raster representation).
Topological relations are based on the boundaries, interiors and complements of the two related
objects and are invariant under transformations which are continuous, one-one, onto and whose inverse is continuous. The relations are: A disjoint B, A meets B, A overlaps B, A equals B, A covers
-3-

C north A
A disjoint B

B northeast A

A overlap B

B
A

A distance=c B

rep(A)

A distance<c B

Figure 7.2. Illustration of some topological, some distance and some direction relations
B, A covered-by B, A contains B, A inside B. A formal definition has been given by Egenhofer
(1991).
Distance relations compare the distance of two objects with a given constant using one of the
arithmetic comparison operators. If dist is a distance function, is one of the arithmetic predicates
<, > or = , and c is a real number, then a distance relation O1 distance c O2 between the two spatial
objects O1 and O2 holds if distance(O1, O2) c.
To define the direction relations, e.g. O2 south O1, we consider one representative point of the
object O1 as the origin of a virtual coordinate system whose quadrants and half-planes define the
directions. To fulfil the direction predicate, all points of O2 have to be located in the respective area
of the plane. Figure 7.2 illustrates the definition of some direction relations using 2D polygons.
Obviously, the directions are not uniquely defined but there is always a smallest direction relation for two objects A and B, called the exact direction relation of A and B, which is uniquely determined. In figure 7.2, for instance, A and B satisfy the direction relations northeast and east but
the exact direction relation of A and B is northeast.
By combining basic spatial relations via logical operators it is possible to define more complex
spatial relations, e.g. O1 being north of O2 and no more than 5 km away. Each such spatial relation induces a spatial neighbourhood graph as defined in the following definition.

-4-

Definition 1: (neighbourhood graphs and paths)

Let neighbour be a neighbourhood relation and DB be a database of spatial objects. A neighbourhood graph

G neighbor = ( N, E )
DB

is a graph with nodes N = DB and edges E N N where an

edge e = (n1, n2) exists iff neighbour(n1,n2) holds. A neighbourhood path of length k is defined as
a sequence of nodes [n1, n2, . . ., nk], where neighbour(ni, ni+1) holds for all n i N, 1 i < k .
We assume the standard operations from relational algebra like selection, union, intersection
and difference to be available for sets of objects and sets of neighbourhood paths (e.g., the operation selection(set, predicate) returns the set of all elements of a set satisfying the predicate predicate). In addition, we introduce some operations which are specific to neighbourhood graphs and
paths and which are designed to support spatial data mining. The following operations are briefly
described:

neighbours: Graphs Objects Predicates Sets_of_objects

paths: Sets_of_objects Sets_of_paths;

extensions: Graphs Sets_of_paths Integer Predicates Sets_of_paths

The operation neighbours(graph, object, predicate) returns the set of all objects connected to

object in graph satisfying the conditions expressed by the predicate predicate.

The operation paths(objects) creates all paths of length 1 formed by a single element of objects
and the operation extensions(graph, paths, length, predicate) returns the set of all paths of the specified length in graph extending one of the elements of paths. The extended paths must satisfy the
predicate predicate. The elements of paths are not contained in the result implying that an empty
result indicates that none of the elements of paths could be extended.
Because the number of neighbourhood paths may become very large, the argument predicate in
the operations neighbours and extensions acts as a filter to restrict the number of neighbours and
paths to certain types of neighbours or paths. The definition of predicate may use spatial as well as
non-spatial attributes of the objects or paths.
For the purpose of KDD, we are mostly interested in paths leading away from the start object.
We conjecture that a spatial KDD algorithm using a set of paths which are crossing the space in
arbitrary ways will not produce useful patterns. The reason is that spatial patterns are most often
the effect of some kind of influence of an object on other objects in its neighbourhood. Furthermore, this influence typically decreases or increases more or less continuously with increasing or
-5-

starlike

variable starlike

vertical starlike

Figure 7.3. Illustration of some filter predicates

decreasing distance. To create only relevant paths, we introduce special filter predicates which
select only particular subsets of all paths, i.e. paths which are leading away from the start objects
in a certain sense. This approach also significantly reduces the runtime of the spatial data mining
algorithms operating on neighbourhood paths.
There are different possibilities to define filters for paths leading away from start objects. The
filter starlike, e.g., is a very restrictive filter which allows only a small number of coarse paths.
This filter is appropriate for many applications, and in terms of runtime it is most efficient. It requires that, when extending a path p = [n1,n2,...,nk] with a node nk+1, the exact final direction of
p may not be generalized. For instance, a path with final direction northeast can only be extended by
a node of an edge with the exact direction northeast. The filter variable-starlike allows more finegrained paths by requiring only that, when extending p the edge (nk, nk+1) has to fulfil at least the
exact initial direction of p. For instance, a neighbourhood path with initial direction north can be
extended such that the direction north or the more special direction northeast is satisfied. The filter
variable-starlike allows a more detailed spatial analysis than the filter starlike, but it increases the
runtime of a data mining algorithm because more paths have to be processed by the algorithm.
Figure 7.3 illustrates these filters when extending the paths from a given start object. The figure
also depicts another filter vertical starlike which is less restrictive in vertical than in horizontal direction. This filter is appropriate when the vertical direction should be analyzed in greater detail
than the horizontal direction.
2.2 Efficient DBMS Support
In this section, we present techniques to efficiently support the proposed database primitives by
a DBMS. Thus, our basic operations for spatial data mining can be integrated into commercial
DBMSs and all the benefits of these systems can be exploited for spatial data mining applications.
-6-

Object-ID

Neighbour

Distance

Direction

Topology

o1

o2

2.7

southwest

disjoint

o1

o3

northwest

overlap

...

...

...

...

...

B+ tree

Figure 7.4. Sample Neighbourhood Index

Typically, spatial index structures, e.g. R-trees (Guttman, 1984), are used in an DBMS to speed
up the processing of spatial queries such as region queries or nearest neighbour queries (Gueting,
1994). If the spatial objects are fairly complex, however, retrieving the neighbours of some object
this way is still very time consuming due to the complexity of the evaluation of neighbourhood relations on such objects. Furthermore, when creating all neighbourhood paths with a given source
object, a very large number of neighbours operations have to be performed. Finally, many spatial databases are rather static since there are not many updates on objects such as geographic maps
or proteins. Therefore, materializing the relevant neighbourhood graphs and avoiding access to the
spatial objects themselves may be worthwhile. Our concept of neighbourhood indices is related to
that of the distance associated join indices (Lu and Han, 1992) with the following new contributions:
A specified maximum distance restricts the pairs of objects represented in a neighbourhood index.
For each of the different types of neighbourhood relations (that is distance, direction, and topological relations), the concrete relation of the pair of objects is stored.
In the following, we introduce neighbourhood indices more formally.
Definition 2: (neighbourhood index)
Let DB be a set of spatial objects and let max and dist be real numbers. Let D be a direction
relation and T be a topological relation. Then the neighbourhood index for DB with maximum disDB

tance max, denoted by I max , is defined as follows:

I max = {(O1,O2,dist,D,T) | O1, O2 DB, O1 distance=dist O2 dist max O2 D O1 O1 T O2}.
DB

A simple implementation of a neighbourhood index using a B+-tree on the key attribute ObjectID is illustrated in figure 7.4.
-7-

A neighbourhood index supports not only one but a set of neighbourhood graphs. We call a
neighbourhood index applicable for a given neighbourhood graph if the index contains an entry for
each of the edges of the graph. To find the neighbourhood indices applicable for some neighbourhood graph, we introduce the notion of the critical distance of a neighbourhood relation r. Intuitively, the critical distance of a neighbourhood relation r is the maximum possible distance for a pair
of objects O1 and O2 satisfying O1 r O2. The following lemma allows us to calculate the critical
distance for any neighbourhood relation. The critical distance is calculated recursively along the
composition of a neighbourhood relation.
Lemma 1: The following equation holds for the critical distance of a neighbourhood relation r:

0 if r is a topological relation except disjoint

d if r is the relation distance<d or distance=d

c-distance(r) =
if r is a direction relation, the relation distance>d
or disjoint
min ( c-distance ( r 1 ), c-distance ( r 2 ) ) if r = r1 r2

max ( c-distance ( r 1 ), c-distance ( r 2 ) ) if r = r1 r2

A neighbourhood index with a maximum distance of max is applicable for a neighbourhood
graph with relation r if the critical distance of r is not larger than max. Obviously, if two neighDB

DB

DB

bourhood indices I c1 and I c2 with c1 < c2 are available and applicable, using I c1 is more efficient
DB

because in general it has less entries than I c2 . The smallest applicable neighbourhood index for
some neighbourhood graph is the applicable neighbourhood index with the smallest critical distance.
In figure 5, we sketch the algorithm for processing the neighbours operation which makes use
of the smallest applicable neighbourhood index. The first step, the index selection, selects a neighbourhood index. If there is no applicable neighbourhood index, then the standard approach of using
some spatial index structure is followed. The filter step returns a set of candidate objects (which
may satisfy the specified neighbourhood relation) with a cardinality significantly smaller than the
database size. In the last step, the refinement step, for all these candidates the neighbourhood relation as well as the additional predicate pred are evaluated and all objects passing this test are returned as the resulting neighbours.
To implement the extensions operation, we perform a depth-first search. Thus, a path buffer of
size O(max-length) is sufficient to store the intermediate results. On the other hand, a breadth-first
-8-

DB

neighbours (graph G r , object o, predicate pred)

DB
select as I the smallest applicable neighbourhood index for G r ; // Index Selection
if such I exists then
// Filter Step
use the neighbourhood index I to retrieve as candidates the set of objects c
having an entry (o,c,dist, D, T) in I
else use the spatial index for DB to retrieve as candidates
the set of objects c satisfying o r c;
initialize an empty set of neighbours;
// Refinement Step
for each c in candidates do
if o r c and pred(c) then
add c to neighbours
return neighbours;
Figure 7.5. Algorithm neighbours
search would require a much larger buffer size since it begins creating all paths before finishing
the first one. To retrieve the nodes for potential extensions of a candidate path, the neighbours operations is used indicating that the efficiency of this operation is crucial.
DB

To create a neighbourhood index I max , a spatial join on DB with respect to the neighbourhood
relation ( O 1 dis tan ce = dist O 2 dist max ) is performed. A spatial join can be efficiently processed by using a spatial index structure. For each pair of objects returned by the spatial join, we
then have to determine the exact distance, the direction relation and the topological relation. The
resulting tuples of the form ( O 1, O 2, Dis tan ce, Direction, Topo log y ) are stored in a relation
which is indexed by a B-tree on the attribute O1.
Updates of a database, i.e. insertions or deletions of spatial objects, require updates of the deDB

rived neighbourhood indices. Fortunately, the update of a neighbourhood index I max is restricted
to the neighbourhood of the respective object defined by the neighbourhood relation A
distance< max B. This neighbourhood can be efficiently retrieved by using either a neighbourhood
index (in case of a deletion) or by using a spatial index structure (in case of an insertion). That
means, it is not required to re-build a neighbourhood index from scratch when an update occurs.
Instead, entries can incrementally be inserted into or deleted from a neighbourhood index.
The database primitives were implemented on top of the commercial DBMS Illustra, Version
3.3, using its 2D spatial data blade which provides R-trees. Neighbourhood indices were realized
in the DBMS by tables and B-trees as described above (see figure 7.4) and the geographic database
-9-

of Bavaria was used for an experimental performance evaluation. The results of our performance
evaluation (Ester et al., 2000) demonstrate a significant speed-up for the neighbours operation with
compared to without neighbourhood index. The neighbours operation is crucial to the efficient
DBMS support of our database primitives since the implementation of the operations for extending
neighbourhood paths is based on the neighbours operation.
In particular, a neighbourhood index is very efficient for complex spatial objects (i.e. large numbers of vertices per spatial object) and an average number of neighbours per spatial object which
is not too large. This is a typical setting, e.g., for geographic information systems when using intersects as the neighbourhood predicate. If the average number of of neighbours is very large for a
spatial neighbourhood relation, we also have to consider the size of the the neighbourhood index.
In the worst case, depending on the critical distance c-distance(r) and the distribution of the spatial
objects, the storage requirements for the materialization of a neighbourhood index can be of O(n2).
Ester et al. (2000) present a cost model for the neighbours operation. Based on system parameters
(e.g. average execution time for a page access) and information about the neigbhorhood relation
(e.g. average number of neighbours) the model can be used to predict the performance gain of a
neighbourhood index and whether this time gain is worth the additional storage requirements.

3 Spatial Clustering
Clustering is the task of grouping the objects of a database into meaningful subclasses (that is,
clusters) so that the members of a cluster are as similar as possible whereas the members of different clusters differ as much as possible from each other. Applications of clustering in spatial databases are, e.g., the detection of seismic faults by grouping the entries of an earthquake catalog or
the creation of thematic maps in geographic information systems by clustering feature vectors. We
can support clustering algorithms by our database primitives if the clustering algorithm is based on
a local cluster condition, i.e. if it constructs clusters by analyzing a restricted neighbourhood of
the objects. Examples are the density-based clustering algorithm DBSCAN (Ester et al., 1996) as
well as its generalized version GDBSCAN (Sander et al., 1998) which is discussed in the following.

- 10 -

3.1 Generalized Density-Based Clustering

GDBSCAN (Generalized Density Based Spatial Clustering of Applications with Noise) relies on
a density-based notion of clusters. The key idea of a density-based cluster is that for each point of
a cluster its Eps-neighbourhood for some given Eps > 0 has to contain at least a minimum number
of points, i.e. the density in the Eps-neighbourhood of points has to exceed some threshold.
Density-based clusters can be generalized to density-connected sets in the following way:
First, any notion of a neighbourhood can be used instead of an Eps-neighbourhood if the definition
of the neighbourhood is based on a binary predicate NPred which is symmetric and reflexive. Second, instead of simply counting the objects in a neighbourhood of an object, other measures to define an eqivalent of the cardinality of that neighbourhood can be used as well. For that purpose
we assume a predicate MinWeight which is defined for sets of objects and which is true for a neighbourhood if the neighbourhood has the minimum weight (e.g. a minimum cardinality as for density-based clusters).
Whereas a distance-based neighbourhood is a natural notion of a neighbourhood for point objects, it may be more appropriate to use topological relations such as intersects or meets to cluster
spatially extended objects such as a set of polygons of largely differing sizes. There are also specializations equivalent to simple forms of region growing (Niemann, 1990), i.e. only local criteria
for expanding a region can be defined by the weighted cardinality function. For instance, the neighbourhood may be given simply by the neighbouring cells in a grid and the weighted cardinality
function may be some aggregation of the non-spatial attribute values. While region growing algorithms are highly specialized to pixels, density-connected sets can be defined for any data types.
The neighbourhood predicate NPred and the MinWeight predicate for these specializations are listed below and illustrated in figure 7.6 (see (Sander et al., 1998) for a more detailed discussion of
neighbourhood relations for different applications):
Density-based clusters: NPred: distance Eps, wCard: cardinality, MinWeight(N): |N| MinPts
Clustering of polygons: NPred: intersects or meets, wCard: sum of areas, MinWeight(N):
sum of areas MinArea

- 11 -

Density-based clusters

Clustering of polygons

Simple region growing

Figure 7.6. Different specializations of density-connected sets

GDBSCAN (DB, NPred, MinWeight)
// Precond.: for all objects o in DB: o.ClId = UNCLASSIFIED and o.Processed = FALSE
NPredGraph := createNeighbourhoodGraph(DB, NPred)
ClusterId := nextId(NOISE);
for i from 1 to DB.size do
Object := DB.get(i);
if not Object.Processed then
if ExpandCluster(Object,ClusterId,NPredGraph,MinWeight) then
ClusterId := nextId(ClusterId)
Figure 7.7. Algorithm GDBSCAN
Simple region growing: NPred: neighbour and similar non-spatial attributes,
MinWeight(N): aggr(non-spatial values) threshold.
All instances of density-connected sets can be constructed by the same algorithm which is based
DB

on a neighbourhood graph G NPred and which uses the neighbours operation introduced above in
our framework for spatial data mining. To find a density-connected set, GDBSCAN starts with an
arbitrary object p and retrieves all objects density-reachable from p with respect to NPred and
MinWeight. Density-reachable objects are retrieved by performing successive NPred-neighbourhood queries and checking the minimum weight of the respective results. If p is a core object, this
procedure yields a density-connected set with respect to NPred and MinWeight. If p is not a core
object, no objects are density-reachable from p and p is assigned to NOISE. This procedure is iteratively applied to each object p which has not yet been classified. Thus, a density-based decomposition of a dataset is constructed.
In figure 7.7, we present the basic version of the algorithm GDBSCAN based on a neighbourhood graph with respect to the neighbourhood predicate NPred:

- 12 -

DB

ExpandCluster(Object, ClId, G NPred , MinWeight):Boolean

DB
neighbours := neighbours( G NPred , Object, true);
Object.Processed := true;
if MinWeight(neighbours) then // object is a core object
Seeds.init(NPred, MinWeight, ClId);
Seeds.update(neighbours, Object);
while not Seeds.empty() do
currentObject := Seeds.next();
DB
neighbours := neighbours( G NPred , currentObject, true);
currentObject.Processed := true;
if MinWeight(neighbours) then
Seeds.update(neighbours, currentObject);
return true;
else // Object is NOT a core object
changeClusterId(Object, NOISE);
return false;
Figure 7.8. Function ExpandCluster
SetOfObjects is either the whole database or a discovered cluster from a previous run. NPred
and MinWeight are the global density parameters. The cluster identifiers are from an ordered and
countable data-type (e.g. implemented by Integers) where UNCLASSIFIED < NOISE < other
Ids, and each object will be marked with a cluster-id Object.ClId. The function nextId(clusterId)
returns the successor of clusterId in the ordering of the data-type (e.g. implemented as Id := Id+1).
The function SetOfObjects.get(i) returns the i-th element of SetOfObjects. In figure 7.8, function
ExpandCluster constructing a density-connected set for a core object Object is presented.
A call of neighbours(NPredGraph, Object, true) returns the NPred-neighbourhood of Object
by using the neighbourhood graph NPredGraph. If the NPred-neighbourhood of Object has minimum weight, the objects from this NPred-neighbourhood are inserted into the set Seeds and the
function ExpandCluster successively performs NPred-neighbourhood queries for each object in
Seeds, thus finding all objects that are density-reachable from Object, i.e. constructing the densityconnected set that contains the core object Object.
The class Seeds controls the main loop of ExpandCluster. The method Seeds.next() selects the
next element from the set Seeds and deletes it from Seeds. The method Seeds.update(neighbos,
centreObject) inserts into the class Seeds all objects from the set neighbours which have not yet
been considered, i.e. which have not already been found to belong to the current density-connected

- 13 -

surface of
the earth

(12),(17.5)




(8.5),(18.7)

Channel 1
Cluster 1
Cluster 2

12

10

1112
1122
3232
3333

feature
space

Cluster 3
8
16.5 18.0 20.0 22.0 Channel 2

Figure 7.9. Relation between 2D image and feature space

set. This method also calls the method to change the cluster-id of the objects to the current clusterId.
3.2 Applications
Application 1: Earth Science (5D points)
In this application, we use a 5-dimensional feature space obtained from several satellite images of a
region on the surface of the earth covering California. These images are taken from the raster data
of the SEQUOIA 2000 Storage Benchmark. After some preprocessing, five images containing
1,024,000 intensity values (8 bit pixels) for 5 different spectral channels for the same region were
combined. Each pixel corresponds to an earth surface area of 1,000 by 1,000 meters and a 5-dimensional feature vector is assigned to this area.
Finding clusters in such feature spaces is a common task in remote sensing digital image analysis
(Richards, 1983) for the creation of thematic maps in geographic information systems. The assumption is that feature vectors for areas of the same type of underground on the earth are forming
groups in the high dimensional feature space (see figure 7.9 illustrating the case of 2D raster images).
We used dist(X,Y) < 1.42 as NPred(X,Y) and cardinality(N) 20 as MinWeight(N). After
some postprocessing we obtained 9 clusters with sizes ranging from 598,863 to 2,016 points. The
postprocessing consisted of two steps: 1. rejecting clusters containing less than 200 points and 2. reassigning the points from the rejected clusters and the noise points to the remaining clusters because
a non-noise class label for each raster point is required for this application.

- 14 -

The result is shown in figure 7.10. Each cluster was coded by a different color. Then each point
in the image of the surface of the earth was colored according to the identificator of the cluster containing the corresponding 5-dimensional vector. A high degree of correspondence between the obtained image and a physical map of California can easily be seen. A detailed discussion of this correspondence is beyond the scope of this chapter.
This is a simplified description of the application. In practice, to obtain high quality results,
there are several pre- and post-processing steps to the clustering of the feature vectors. Especially,
the clustering of the feature vectors should be followed by a spatial smoothing because feature vectors which are close to each other in space are also likely to belong to the same class.
Application 2: Geography (2D polygons)
In the following, we present a simple method for detecting influence regions in a geographic database. GDBSCAN is used to extract density-connected sets of neighbouring objects having a similar value of the non-spatial attribute(s). To define the similarity on an attribute, its domain is partitioned into a number of disjoint classes and values in the same class are considered similar to each
other. The sets with the highest or lowest attribute value(s) are most interesting and are called influence regions, i.e. the maximal neighbourhood of a centre having a similar value in the non-spatial
attribute(s) as the centre itself. For economic geography, the resulting influence region may be further analyzed by comparing them to a circular region representing the expected theoretical shape to
obtain a possible deviation. Different methods may be used for this comparison. A difference-based
method calculates the difference of both, the observed influence region and the theoretical circular
region, thus returning some region indicating the location of a possible deviation. An approxima-

Figure 7.10. Visualization of the clustering result for the SEQUIOA 2000 raster data
- 15 -

tion-based method calculates the optimal approximating ellipsoid of the observed influence region.
If the two main axes of the ellipsoid differ in length significantly, then the longer one is returned indicating the direction of a deviation. Ester et al. (1997) present a detailed description of this application.
GDBSCAN can be used to extract the influence regions from a spatial database. We define
NPred(X,Y) as intersect(X,Y) attr-class(X) = attr-class(Y) and use cardinality as wCard function. Furthermore, we set MinCard to 2 in order to exclude sets of less than 2 objects. Some results
of this approach for the BAVARIA database are illustrated in figure 7.11 .

4 Spatial Characterization
The task of characterization is to find a compact description for a selected subset of the database. In this section, we discuss the task of characterization in the context of spatial databases and
present two relevant methods.
4.1 Algorithms
The task of mining association rules has been introduced by Agrawal and Srikant (1994). An
association rule is a rule I1 I2 where I1 and I2 are disjoint sets of items. The support of the rule
is given by the number of database tuples containing all elements of I1 and the confidence is given
by the number of tuples containing all elements of both I1 and I2. For a database DB of transactions
(i.e. records contain sets of items bought by some customer in one transaction), all association rules
should be discovered having a support of at least minsupp and a confidence of at least minconf in
DB.

Figure 7.11. Illustration of the influence regions of Ingolstadt (left) and Munich (right)
and their deviation from the expected shape
- 16 -

Extending the general concept of association rules, Koperski and Han (1995) introduce spatial
association rules which describe associations between objects based on spatial neighbourhood relations. For instance, a user may want to discover the spatial associations of towns in British Columbia with roads, waters or boundaries having some specified support and confidence.
Figure 7.12 depicts the specification of this spatial data mining task.
discover spatial association rules
inside British_Columbia
from road R, water W, mines M, boundary B
in relevance to town T
where close-to(T.geo, X.geo) and X in {R, W, M, B}
having minsupp = 5 % and minconf = 80 %
Figure 7.12. Example specification for mining spatial association rules
Then, the following spatial association rule may be discovered:
X DB Y DB: is-a(X,town) close-to(X,Y) is-a(Y,water) (80%)
This rule states that 80% of the selected towns are close to water, i.e. the rule characterizes
towns in British Columbia as generally being close to some lake, river etc.
The input for mining spatial association rules specifies a relation of n tuples with a spatial attribute, a spatial neighbourhood relation, a concept hierarchy for each of the attributes, a selection
of relevant object types, the minimum support, and the mininmum confidence.

The proposed algorithm consists of five steps. Step 2 (coarse spatial computation) and step
4 (refined spatial computation) involve spatial aspects of the objects and thus are examined in
the following. Step 2 computes spatial joins of the object type to be characterized (such as
town) with each of the other specified object types (such as water, road, boundary or mine) using the neighbourhood relation (such as close-to). For each of the candidates obtained from
step 2 which passed step 3, in step 4 the exact spatial relation, for example overlap, is determined. Finally, a relation such as the one depicted in figure 7.13 results which is the input for the
final, non-spatial, Apriori-like step of rule generation. To implement this algorithm using our database primitives it is sufficient to replace step 2 by the following procedure. The spatial join can
be replaced by calling a neighbours operation for each target object selected in step 1. The under-

- 17 -

lying neighbourhood graph in this case is defined by the user-specified neighbourhood relation
(e.g. close-to).
Town
Saanich

Water
<meet, J.FucaStrait>

PrinceGeorge
Petincton
...

Road
<overlap,highway1>,
<close-to,highway17>

Boundary
<close-to,US>

<overlap, highway97>
<meet,OkanaganLake>

<overlap, highway97>

...

...

<close-to,US>
...

Figure 7.13. Input for the step of rule generation

As shown by Koperski and Han (1995), the proposed algorithm can be efficiently implemented,
and it is included in the GeoMiner system (Han et al., 1997). A limitation of spatial association
rules is that they do not use the non-spatial attributes to characterize the specified objects.
We define a spatial characterization of a set of target objects with respect to the database containing them as a description of the spatial and non-spatial properties which are typical for the target objects but not for the whole database. We use the relative frequencies of the non-spatial attribute values and the relative frequencies of the different object types as the interesting properties.
To obtain a spatial characterization, we consider not only the properties of the target objects,
but also the properties of their neighbours up to a given maximum number of edges in the neighbourhood graph. Figure 7.14 illustrates relative frequencies in the database as well as in the target
regions and the ratio of these frequencies in comparison with the specified level of significance.
We define a property of some spatial object by a tuple of one of the two following formats: either (attribute, value) or (type, type) where type denotes the special attribute representing
the object type such as community, mountain or river. The task of spatial characterization is to discover the set of all properties for which the relative frequency in a set targets, extended by its
neighbours, is significantly different from the relative frequency in DB. A very frequent property
present only in the neighbourhood of very few of the targets would create misleading results.

- 18 -

Figure 7.14. Sample frequencies and ratios

Therefore, we also require that such a property must have a significantly different, either smaller
or larger, relative frequency in the neighbourhood of at least proportion many targets.
DB

Definition 3: (spatial characterization): Let G neighbor be a neighbourhood graph and targets be

a subset of DB. Let freqs(prop) denote the number of occurrences of the property prop in the set s
and let card(s) denote the cardinality of s. The frequency factor of prop with respect to targets and
DB, denoted by f t arg ets ( prop ) , is defined as follows:
DB

t arg ets
DB
DB
freq
( prop )- freq
( prop )
f t arg ets ( prop ) = --------------------------------------- ----------------------------------

card ( t arg ets )

card ( DB )

Let significance and proportion be real numbers and let max-neighbours be a natural number.
Let neighbors Gi ( s ) denote the set of all objects reachable from one of the elements of s by traversing at most i of the edges of the neighbourhood graph G. Then, the task of spatial characterization
is to discover each property prop and each natural number n max-neighbours such that (1) the set
objects = neighborsG ( targets ) as well as (2) the sets objects = neighbors G ( { t } ) for at least pron

portion many t targets satisfy the condition:

f objects ( prop ) significance
DB

or

1
DB
f objects -------------------------------significance

In point (1) the union of the neighbourhood of all target objects is considered simultaneously,
whereas in point (2) the neighbourhood of each targets is considered separately. The parameter
proportion specifies the minimum confidence required for the characterization rules and the frequency factors of the properties provide a measure of their interestingness with respect to the target
objects.
Figure 7.15 presents the algorithm for discovering spatial characterizations. The parameter proportion is relevant only for the last step of the algorithm, i.e. for the generation of a rule. Note the
- 19 -

DB

discover-spatial-characterization(graph G r ; set of objects targets;

real significance, proportion; integer max-neighbours)
initialize the set of characterizations as empty;
initialize the set of regions to targets;
initialize n to 0;
calculate frequencyDB(prop) for all properties prop = (attribute, value) in DB;
while n max-neighbours do
for each attribute of DB and for the special attribute object type do
for each value of attribute do
calculate frequencyregions(prop) for property prop = (attribute, value);
DB
DB
if f regions
( prop ) significance or f regions
( prop ) 1 / significance then
DB
add (prop, n, f regions
( prop ) ) to the set characterizations;
if n < max-neighbours then
for each object in regions do
DB
add neighbours( G r , object, true) to regions;
increment n by 1;
extract all tuples (prop, n, factor) from characterizations which are significant in at
least proportion of the regions with n extensions;
return the rule generated from these characterizations;
Figure 7.15. Algorithm discover-spatial-characterization
importance of the parameter max-neighbours (that is, the maximum number of edges of the neighbourhood graph traversed starting from a target object) in the resulting characterizations. For example, a property may be significant when considering all neighbours which are reachable from
one of the target objects via 2 edges of the neighbourhood graph. However, the same property may
not be significant when considering further neighbours if then the target regions are extended by
objects for which the property is not frequent. The generated rule has the following format:
target p1 (n1, freq-fac1) ... pk (nk, freq- fack).
This rule means that for the set of all targets extended by ni neighbours, the property pi is freqfaci times more (or less) frequent than in the database.
4.2 Application: Economic Geography
A geographic database may be used, e.g., by economic geographers to discover spatial characterizations of the economic power or other interesting properties of communities. Some non-spatial
attribute such as the unemployment rate is chosen as a relevant indicator.
The BAVARIA database was used for an experimental performance evaluation of the algo-

rithms (Ester et al., 1998).

- 20 -

(a) target objects

(b) final target regions

Figure 7.16. Characterization wrt. high proportion of retired persons

The characterization algorithm usually starts with a small set of target objects, selected for instance by a condition on some non-spatial attribute(s) such as proportion of retired persons =
HIGH (see figure 7.16 (a)). Then, the algorithms expands regions around the target objects, simultaneously selecting those attributes of the regions for which the distribution of values differs
significantly from the distribution in the whole database (figure 7.16 (b)).
In the last step of the algorithm, a characterization rule is generated describing the target regions
(figure 7.17). Informally, this rule states that retirees prefer somewhat rural areas close to the
mountains. Note that not only some non-spatial attributes but also the neighbourhood of mountains (after three extensions) are significant for the characterization of the target regions.
community has a high proportion of retired persons
appartments per building =

very low (n = 0, f(prop) = 9.1)

rate of foreigners =

very low (n = 0, f(prop) = 8.9)

rate of academics =

medium (n = 0, f(prop) = 6.3)

average size of enterprises = very low (n = 0, f(prop) = 5.8)

object type = mountain (n = 3, f(prop) = 4.1)
Figure 7.17. Rules characterizing the target objects of figure 7.16

- 21 -

.
.
.
.. .... ..
.

difference

difference

difference

..
. .. .. .
.. . .

distance

distance

(a) positive trend

(b) negative trend

.. ...
.. ....
.

distance

(c) no trend

Figure 7.18. Sample linear trends

5 Spatial Trend Detection

We define a spatial trend as a regular change of one or more non-spatial attributes when moving
away from a given start object o. We use neighbourhood paths starting from o to model the movement and we perform a regression analysis on the respective attribute values for the objects of a
neighbourhood path to describe the regularity of change.
5.1 Algorithms
First, we define the task of spatial trend detection for some source object o1 DB. To detect
regular changes of some non-spatial attributes, we perform a regression analysis as follows. The
independent variable (X) yields the distance between any database object o2 and the source object.
The dependent variable (Y) measures the difference of the values of some non-spatial attribute(s)
for o1 and o2. Then, the sets X and Y contain one observation for each element of a subset S of DB.
If the absolute value of the correlation coefficient is found to be large enough, S identifies a part of
DB showing a significant spatial trend for the specified attributes(s) starting from o1. In the following, we will use linear regression, since it is efficient and often the influence of some phenomenon
to its neighbourhood is either linear or may be transformed into a linear model, e.g. exponential
regression. Figure 7.18 illustrates a strong positive (correlation coefficient >> 0) and a strong negative (correlation coefficient << 0) linear trend as well as a situation where no significant (linear)
trend is observed.

- 22 -

In a naive approach, one observation could be made for each element o2 DB and a regression
analysis would be conducted once for this whole set of observations. This approach, however,
would fail to discover a spatial trend in the following situations:

a trend is present only locally in the neighbourhood but not globally in the database
a trend is present not in all directions but only in some directions.
We argue that such situations are very common in real databases. We use the concept of neighbourhood paths to overcome these problems and consider only such objects o2 which are located
on one of the neighbourhood paths starting from o1. We stop extending a neighbourhood path as
soon as no more significant trend can be found and, thus, provide a means to restrict the search to
the neighbourhood of the source object. Furthermore, it is enough that some, not all, neighbourhood paths show a spatial trend of the specified type.
Definition 4: (spatial trend detection): Let g be a neighbourhood graph, o an object (node) in g
and a be a subset of all non-spatial attributes. Let t be a type of function, e.g. linear or exponential,
used for the regression and let filter be one of the filters for neighbourhood paths. Let min-conf be
a real number and let min-length as well as max-length be natural numbers. The task of spatial
trend detection is to discover the set of all neighbourhood paths in g starting from o and having a
trend of type t in attributes a with a correlation of at least min-conf. The paths have to satisfy the
filter and their length must be between min-length and max-length.
Definition 7.4 allows different specializations. Either the set of all discovered neighbourhood
paths or each of its elements must have a trend of the specified type. For each of these specializations, we present an algorithm to discover such spatial trends. Both algorithms require the same
input parameters but they use different methods to search the set of all relevant neighbourhood
paths.
The first algorithm discovers global trends, i.e. trends for the whole set of all neighbourhood
paths with source o having a length in the specified interval. Algorithm detect-global-trends performs a breadth-first search of the set of all neighbourhood paths starting from o and it is outlined
in figure 7.19.
Beginning from o, the algorithm creates all neighbourhood paths of the same length simultaneaously - starting with min-length and continuing until max-length. The regression is performed
- 23 -

DB

detect-global-trends(graph G r ; object o; attribute a; type t; real min-conf,

integer min-length,max-length; filter f)
DB
initialize a list of paths to the set extensions( G r , {path(o)}, min-length, f);
initialize an empty set of observations;
initialize last-correlation and last-paths as NULL;
initialize first-pos to 1;
initialize last-pos to min-length;
while paths is not empty do
for each path in paths do
for object from first-pos of path to last-pos of path do
calculate diff as a(object) - a(o) and calculate dist as dist(object,o);
insert the tuple (diff, dist) into the set of observations;
perform a regression of type t on the set of observations;
if abs(correlation) of the resulting regression function min-conf then
set last-paths to paths and last-correlation to correlation;
set last-slope to the slope of the resulting regression function;
if length(paths)< max-length then
DB
replace the paths by the set extensions( G r , paths, length(paths) + 1, f);
increment last-pos by 1;
set first-pos to last-pos;
else set paths to the empty list;
else return last-paths, last-correlation and last-slope;
return last-paths, last-correlation and last-slope;
Figure 7.19. Algorithm detect-global-trends
once for each of these sets of all paths of the same length. If no trend of length l with abs(correlation) min-conf is detected, then the path extensions of length l+1, l+2, . . ., max-length are not
created due to efficiency reasons. The parameter min-length has to be chosen large enough because
a trend may become significant only at a certain minimum length of a neighbourhood path. The
algorithm returns the set of neighbourhood paths of the maximum length having a trend with
abs(correlation) min-conf. Furthermore, the slope of the regression function and its correlation
coefficient are returned. The slope describes the degree of change of the specified attribute when
moving away from the source object o: a large slope is found for a quickly changing attribute
whereas a small slope indicates that the attribute changes only slowly in the neighbourhood of o.
The correlation coefficient measures the confidence of the discovered trend - the larger the correlation coefficient the better can the obervations be approximated by the regression function and the
higher is the confidence of the discovered trend.

- 24 -

The second algorithm discovers local trends, i.e. trends for a single neighbourhood path with
source o having a length in the specified interval. Algorithm detect-local-trends creates the neighbourhood paths starting from o in a depth-first manner which is presented in figure 7.20. A regression is performed once for each of the neighbourhood paths with max-length length min-length
and a path is only extended further if it has a significant trend (i.e. abs(correlation) of the resulting
regression function min-conf). The algorithm returns two sets of paths showing a significant spatial trend: a first set of positive trends and a second set of negative trends.
DB

detect-local-trends(graph G r ; object o; attribute a; type t; real min-conf,

integer min-length,max-length; filter f)
DB
initialize a list of paths to the set extensions( G r , path(o), min-length, f);
initialize two empty sets of positive-trends and negative-trends;
while paths is not empty do
initialize the set of observations as empty;
remove the next element of paths and take it as path;
for object from min-length-th object of path to last object of path do
calculate diff as a(object) - a(o) and calculate dist as dist(object,o);
insert the tuple (diff,dist) into the set of observations;
perform a regression of type t on the set of observations;
if abs(correlation) of the resulting regression function min-conf then
if correlation > 0 then
insert the tuple (path, correlation) into the set of positive-trends;
else
insert the tuple (path, correlation) into the set of negative-trends;
if length(path) < max-length then
DB
add the extensions( G r , path, length(path) + 1, f) to the head of paths;
return positive-trends and negative-trends;
Figure 7.20. Algorithm detect-local-trends
5.2 Application: Economic Geography
Economic geographers may be interested in discovering spatial trends of various properties of
communities such as the economic power. In the following, we illustrate the potential of spatial
trend detection on the BAVARIA database which was used for an experimental performance eval-

uation of the algorithms (Ester et al., 1998).

Spatial trends describe a regular change of non-spatial attributes when moving away from a
start object o. The two above algorithms may produce different patterns of change for the same start
object o.

- 25 -

The existence of a global trend for a start object o indicates that if considering all objects on all
paths starting from o the values for the specified attribute(s) in general tend to increase (decrease)
with increasing distance. Figure 7.21 (a) depicts the result of algorithm global-trend for the attribute average rent and the city of Regensburg as a start object.
direction of decreasing values of trend attribute average rent

Regensburg

(b) local-trends

(a) global-trends

Figure 7.21. Visualization of trends starting from the city of Regensburg

Algorithm local-trends detects single paths starting from an object o and having a certain trend.
The paths starting from o may show different pattern of change, e.g., some trends may be positive
while the others may be negative. Figure 7.21 (b) illustrates this case for the attribute average
rent and the city of Regensburg as a start object.
Both algorithms were applied to the Bavaria database varying min-confidence from 0.6 to 0.8
for the attribute average rent and linear type of regression. The predicate intersects was used as
the neighbourhood relation to define the graph. The filter vertical starlike for paths was used because due to our domain knowledge we expected the most significant trends in north-south direction. The length of the paths was restricted by min-length = 4 and max-length = 7.
The spatial objects within a trend region, i.e. either the start objects or the objects forming the
paths, may be the subject of further analysis. For instance, algorithm global-trend may detect regions showing a certain global trend, and algorithm local-trends then finds within these regions
some paths having the inverse trend (see figure 7.21). Then, we may try to find an explanation for
those inverse paths.
Another possibility of further analysis of the discovered trends is as follows:

- 26 -

First, detect centres for a given attribute, i.e. communities with a significantly decreasing
spatial trend of this attribute. Algorithm global-trend can be used for this task.

Second, apply the spatial characterization to these centres to find their common properties.

Ester et al. (1998) report the results of the following experiment. In a first step, centres for
attribute average rent were detected: minimum correlation was set to 0.7 and only those communities were selected where the slope of the trend was less than -10-4 and the path length was not
smaller than 5, i.e. only linear trends that are noticeably decreasing were detected.
With this definition, 24 centres out of the 2043 communities were found. The characterization
rule discovered for these centres contains the following properties:
community is a centre
rate of academics = high (n = 1, f(prop) = 9.1)
average number of persons per household = low (n = 1, f(prop) = 2.5)
rate of foreigners = low (n = 1, f(prop) = 2.8)
Note that none of the attributes was significant for n = 0, i.e. without considering the neighbourhood of the target object. Only if we extend the target regions by one neighbour, we can see characteristic properties. Thus, this result could not be found by a non-spatial characterization algorithm.

6 Spatial Classification
The task of classification is to assign an object to a class from a given set of classes based on
the attribute values of the object. In spatial classification the attribute values of neighbouring objects may also be relevant for the membership of objects and therefore have to be considered as
well.
Ester et al. (1997) proposed a spatial classification algorithm based on the well-known ID3 algorithm (Quinlan, 1986) which was designed for relational databases. The extension to spatial attributes is to consider also the attribute of objects on a neighbourhood path starting from the current
object. Thus, we define generalized attributes for a neighbourhood path p = [o1, . . ., ok] as tuples
(attribute-name, index) where index is a valid position in p representing the attribute with attribute-

- 27 -

name of object oindex. The generalized attribute (economic-power,2), e.g., represents the attribute
economic-power of some (direct) neighbour of object o1.
population of city

medium

low

high
type of neighbour of city

amount of taxes
of city

city
very
low low
. . .

high

road

airport

very
high
economic power
of city = high

type of
neighbour of neighbour of city
airport

(87 %)

road

city

. . .
economic power
of city = high
(95 %)

. . .

IF population of city = low AND amount of taxes of city = very high

THEN economic power of city = high (87 %)
IF population of city = high AND type of neighbour of city = road
AND type of neighbour of neighbour of city = airport
THEN economic power of city = high (95 %)

Figure 7.22. Sample decision tree and rules discovered by the classification algorithm
Because it is reasonable to assume that the influence of neighbouring objects and their attributes
decreases with increasing distance, we can limit the length of the relevant neighbourhood paths by
an input parameter max-length. Furthermore, the classification algorithm allows the input of a
predicate to focus the search for classification rules on the objects of the database fulfilling this
predicate. Figure 7.22 depicts a sample decision tree and two rules derived from it. Economic
power has been chosen as the class attribute and the focus is on all objects of type city.
In figure 7.23, we present the pseudo code of the algorithm for spatial classification which finds
all paths from the root to one of the leaves of a decision tree, where all attributes yield an informa-

- 28 -

tion gain of at least . Note that the algorithm uses a predicate larger_distance, i.e.distance(n1,
ni+1) > distance(n1, ni) to restrict the creation of neighbourhood paths.
spatial_classification_rules(db:Set_of_Objects; class_attr:Attribute;
NPred:NeighbourhoodRelation; max_length:Int)
NPredGraph:= create the neighbourghood graph for db w.r.t. NPred;
paths:=extensions(NPredGraph, {db}, max_length, larger_distance);
classify(class_attr, NPredGraph, EMPTY_RULE, paths, max_length);
classify(class_attr:Attribute; rule:ClassificationRule; paths:set_of_paths; max_length:Int)
max_info_gain:=0.0;
max_attr:=NULL;
for i from 1 to max_length do
for each generalized attribute (Aj,i) not used in rule do
info-gain:=calculate_information_gain(Aj, class_attr, i, paths);
if info_gain > max_info_gain then
max_attr:=Aj;
max_neighbours:=i;
max_info_gain:=info_gain;
if max_attr NULL and max_info_gain > then
for each value of max_attr do
extended_rule:=rule + max_attr,max_neighbours,value;
classify(class_attr, extended_rule, paths, max_length);
calculate_information_gain(attr,class_attr: Attribute; index:Int; paths:set_of_paths);
for each path in paths do
consider attr of the index-th object of path and class_attr of the first object of path
for the calculation of the information gain
Figure 7.23. Algorithm spatial_classification_rules
Another algorithm for spatial classification is presented by Koperski et al. (1998). It works as
follows: The relevant attributes are extracted by comparing the attribute values of the target objects
with the attribute values of their nearest neighbours. The determination of relevant attributes is
based on the concept of the nearest hit (the nearest neighbour belonging to the same class) and the
nearest miss (the nearest neighbour belonging to a different class). In the construction of the decision tree, the neighbours of target objects are not considered individually. Instead, so-called buffers
are created around the target objects and the non-spatial attribute values are aggregated over all objects contained in the buffer. For instance, in the case of shopping malls a buffer may represent the

- 29 -

area where its customers live or work. The size of the buffer yielding the maximum information
gain is chosen and this size is applied to compute the aggregates for all relevant attributes.
Whereas the property of being a nearest neighbour cannot be directly expressed by our neighbourhood relations, it is possible to extend our set of neighbourhood relations accordingly. The
proposed database primitives are, however, sufficient to express the creation of buffers for spatial
classification by using a distance-based neigborhood predicate.

7 Conclusions
In this chapter, we introduced a database-oriented framework for spatial data mining which is
based on the concepts of neighbourhood graphs and paths. A small set of basic operations on these
graphs and paths were defined as database primitives for spatial data mining. Furthermore, techniques to efficiently support the database primitives by a commercial DBMS were presented. In
the following sections, we covered the main tasks of spatial data mining: spatial clustering, spatial
characterization, spatial trend detection and spatial classification. For each of these tasks, we presented algorithms as well as prototypical applications in domains such as the earth sciences and
geography. Thus, we demonstrated the practical impact of these algorithms of spatial data mining.
The following issues indicate interesting directions for future research. The database primitives
were implemented on top of the commercial DBMS Illustra. Since the system overhead imposed
by this DBMS is rather large, techniques of improving the efficiency should be investigated. For
example, techniques for processing sets of related neighbours operations which provide more information to the DBMS can be used to improve the overall efficiency of mining algorithms using
the database primitives.
In some spatial databases the dimension of time plays an important role: the history of the relevant part of the world is stored for the purpose of analysis, for example raster images of the same
area of the surface of the earth taken at different times. Data mining in such spatio-temporal databases is a promising area of future research. For example, geographers may be interested in learning spatio-temporal rules describing the process of growth of urban landuse.

- 30 -

References
Agrawal R., Imielinski T., and Swami A. 1993 Database Mining: A Performance
Perspective, IEEE Transactions on Knowledge and Data Engineering, Vol. 5, No. 6, pp. 914925.
Agrawal R., and Srikant R. 1994 Fast Algorithms for Mining Association Rules in Large
Databases Proc. 20th Int. Conf.Very Large Data Bases (VLDB 94), pp. 478-499.
Chen M.-S., Han J., and Yu P. S. 1996 Data Mining: An Overview from a Database
Perspective, IEEE Trans. on Knowledge and Data Engineering, Vol. 8, No. 6, pp. 866-883.
Egenhofer M. J. 1991 Reasoning about Binary Topological Relations, Proc. 2nd Int. Symp.
on Large Spatial Databases, Zurich, Switzerland, pp. 143-160.
Ester M., Frommelt A., Kriegel H.-P., and Sander J. 1998 Algorithms for Characterization and
Trend Detection in Spatial Databases, Proc. 4th Int. Conf. on Knowledge Discovery and Data
Mining, New York City, NY, pp. 44-50.
Ester M., Frommelt A., Kriegel H.-P., and Sander J. 2000 Spatial Data Mining: Database
Primitives, Algorithms and Efficient DBMS Support, Data Mining and Knowledge
Discovery, an International Journal, Kluwer Academic Publishers, Vol. 4, No. 2/3.
Ester M., Kriegel H.-P., and Sander J. 1997 Spatial Data Mining: A Database Approach,
Proc. 5th Int. Symp. on Large Spatial Databases, Berlin, Germany, pp. 47-66.
Ester M., Kriegel H.-P., Sander J., and Xu X. 1996 A Density-Based Algorithm for
Discovering Clusters in Large Spatial Databases with Noise. Proc. 2nd Int. Conf. on
Knowledge Discovery and Data Mining. Portland, Oregon, AAAI Press, Menlo Park,
California, pp. 226-231.
Ester M., Kriegel H.-P., Sander J., and Xu X. 1997: Density-Connected Sets and their
Application for Trend Detection in Spatial Databases. Proc. 3nd Int. Conf. on Knowledge
Discovery and Data Mining. Newport Beach, California, AAAI Press, Menlo Park, California.
Fayyad U. M. 1997 Knowledge Discovery in Databases: An Overview, ILP 1997, pp. 3-16.
Fayyad U. M., Piatetsky-Shapiro G., and Smyth P. 1996 Knowledge Discovery and Data
Mining: Towards a Unifying Framework, Proc. 2nd Int. Conf. on Knowledge Discovery and
Data Mining. Portland, Oregon, AAAI Press, Menlo Park, California, pp. 82 - 88.
Gueting R. H. 1994 An Introduction to Spatial Database Systems, Special Issue on Spatial
Database Systems of the VLDB Journal, Vol. 3, No. 4.
Guttman A. 1984 R-trees: A Dynamic Index Structure for Spatial Searching, Proc. ACM
SIGMOD Int. Conf. on Management of Data, pp. 47-54.
Han J., Koperski K., and N. Stefanovic 1997 GeoMiner: A System Prototype for Spatial Data
Mining, Proc. ACM-SIGMOD Int. Conf. on Management of Data (SIGMOD '97), Tucson,
AR.
- 31 -

Koperski K., Adhikary J., Han J. 1996 Knowledge Discovery in Spatial Databases: Progress
and Challenges, Proc. SIGMOD Workshop on Research Issues in Data Mining and
Knowledge Discovery, Technical Report 96-08, University of British Columbia, Vancouver,
Canada.
Koperski K., and Han J. 1995 Discovery of Spatial Association Rules in Geographic
Information Databases, Proc. 4th Int. Symp. on Large Spatial Databases (SSD 95), Portland,
ME, pp 47-66.
Koperski K., Han J., and Stefanovic N. 1998 An Efficient Two-Step Method for Classification
of Spatial Data, Proc. Symposium on Spatial Data Handling (SDH 98), Vancouver, Canada.
Lu W., and Han J. 1992 Distance-Associated Join Indices for Spatial Range Search, Proc. 8th
Int. Conf. on Data Engineering, Phoenix, AZ, pp. 284-292.
Ng R. T., and Han J. 1994 Efficient and Effective Clustering Methods for Spatial Data
Mining, Proc. 20th Int. Conf. on Very Large Data Bases, Santiago, Chile, 1994, pp. 144-155.
Niemann H. 1990 Pattern Analysis and Understanding. Springer-Verlag, Berlin.
Quinlan J.R. 1986 Induction of Decision Trees, Machine Learning, Vol.1, pp. 81-106.
Richards A.J. 1983. Remote Sensing Digital Image Analysis. An Introduction. Berlin:
Springer Verlag.
Rotem D. 1991 Spatial Join Indices, Proc. 7th Int. Conf. on Data Engineering, Kobe, Japan,
pp. 500-509.
Sander J., Ester M., Kriegel H.-P., and Xu X. 1998 Density-Based Clustering in Spatial
Databases: A New Algorithm and its Applications, Data Mining and Knowledge Discovery,
an International Journal, Kluwer Academic Publishers, Vol.2, No. 2.
Valduriez P. 1987 Join Indices, ACM Transactions on Database Systems, Vol. 12, No. 2,
pp. 218-246.

- 32 -