1 Spatial Data on the Web: Issues and Challenges
7
a set of parameters applicable to a certain class of concepts of a given schema
(e.g. geometric properties, attributes). Among quality parameters, those concerning
imperfections like accuracy, precision, and consistency play an important role.
Accuracy refers to the correspondence between the geographical information and
the real world they represent. Precision deals with the granularity level by which the
real world is represented inside a database, i.e. its resolution. Consistency refers to
the presence of contradictory concepts inside the same data set. Values for quality
parameters can be used during ontology alignment, to improve the quality of the detected spatial relationships, and integration, in order to come up with a global schema
having a certain quality level.
The quality problem can be tackled from the bottom, by extending data integration solutions to cope with quality parameters, or from the top, by extending query
specification and processing to deal with quality issues. One of the key observations is that in the presence of data with different granularities, as in the case of
multiresolution and multiscale data, the specification of equality-based queries, by
which the user specifies in an exact way the constraints that data to be retrieved
must satisfy, may not be the right choice, since multiresolution is not effectively
used during such processing. In order to exploit multiresolution, a possible approach
would be that of introducing some mechanism of query relaxation, by which the
specific characteristics of multiresolution maps are taken into account during query
execution, based on the actual resolution and scale of the data source; as a consequence, approximated answers are returned to the user, possibly introducing some
false hits, but at the same time making query answers more satisfactory from the user
point of view.
Part II of this book consists of three chapters that cover complementary issues in
the context of spatial data integration, according to what was discussed above.
Ontology alignment is the topic of Chap. 6. In particular, after reviewing ontology alignment, ontology integration, and geographical information fusion problems,
the rosetta system is presented, for extensional and automated geographical information fusion based on inductive inference. Discussions concerning how the system
can cope with various types of uncertainty (inaccuracy, imprecision, vagueness) are
also provided.
Chapter 7 deals with quality-enabled spatial mediation systems. Besides presenting the basic problems and the possible solutions, a quality-enabled spatial integration system called VirGIS/Q is presented. VirGIS/Q relies on the existing standards
for geographical data representation and access, as well as for data quality parameters specification.
Chapter 8 presents some qualitative techniques for query relaxation over multiresolution spatial data sets, where different types could be assigned to the same
geographical feature in distinct data sets. The considered information is qualitative
in the sense that it corresponds to qualitative (topological and directional) relationships between spatial objects. The proposed techniques rely on some distance functions, one for each class of relationships, that can be used to relax user queries.
7
a set of parameters applicable to a certain class of concepts of a given schema
(e.g. geometric properties, attributes). Among quality parameters, those concerning
imperfections like accuracy, precision, and consistency play an important role.
Accuracy refers to the correspondence between the geographical information and
the real world they represent. Precision deals with the granularity level by which the
real world is represented inside a database, i.e. its resolution. Consistency refers to
the presence of contradictory concepts inside the same data set. Values for quality
parameters can be used during ontology alignment, to improve the quality of the detected spatial relationships, and integration, in order to come up with a global schema
having a certain quality level.
The quality problem can be tackled from the bottom, by extending data integration solutions to cope with quality parameters, or from the top, by extending query
specification and processing to deal with quality issues. One of the key observations is that in the presence of data with different granularities, as in the case of
multiresolution and multiscale data, the specification of equality-based queries, by
which the user specifies in an exact way the constraints that data to be retrieved
must satisfy, may not be the right choice, since multiresolution is not effectively
used during such processing. In order to exploit multiresolution, a possible approach
would be that of introducing some mechanism of query relaxation, by which the
specific characteristics of multiresolution maps are taken into account during query
execution, based on the actual resolution and scale of the data source; as a consequence, approximated answers are returned to the user, possibly introducing some
false hits, but at the same time making query answers more satisfactory from the user
point of view.
Part II of this book consists of three chapters that cover complementary issues in
the context of spatial data integration, according to what was discussed above.
Ontology alignment is the topic of Chap. 6. In particular, after reviewing ontology alignment, ontology integration, and geographical information fusion problems,
the rosetta system is presented, for extensional and automated geographical information fusion based on inductive inference. Discussions concerning how the system
can cope with various types of uncertainty (inaccuracy, imprecision, vagueness) are
also provided.
Chapter 7 deals with quality-enabled spatial mediation systems. Besides presenting the basic problems and the possible solutions, a quality-enabled spatial integration system called VirGIS/Q is presented. VirGIS/Q relies on the existing standards
for geographical data representation and access, as well as for data quality parameters specification.
Chapter 8 presents some qualitative techniques for query relaxation over multiresolution spatial data sets, where different types could be assigned to the same
geographical feature in distinct data sets. The considered information is qualitative
in the sense that it corresponds to qualitative (topological and directional) relationships between spatial objects. The proposed techniques rely on some distance functions, one for each class of relationships, that can be used to relax user queries.