Geo-Spatial Interoperability: Crossing Semantic and Syntactic barrier in GIS
2. Syntax and Semantics of GIS data
Research in information systems interoperability is motivated by the ever increasing heterogeneity of the computer world. Heterogeneity in GIS is not an exception, but the complexity and richness of geographic data and the difficulty of their representation raise specific issues for GIS interoperability. Besides interoperability has to overcome complexity of sharing and integrating data between systems with different data structures and models, it also has to deal with semantic heterogeneity. The heterogeneity in GIS information may be classified as syntactic heterogeneity, structural heterogeneity, and semantic heterogeneity.
- Syntactic heterogeneity means that various data models are used for data storage and access.
- Structural heterogeneity means that different information systems store their data in different structures.
- Semantic heterogeneity occurs due to the non-uniformity in naming of different GIS entities and their meaning.
Today, research on interoperability solutions is the way to migrate away from the monolithic systems that dominate the GIS market [4].
2.1 Syntactic and Structural Heterogeneity
There are several attempts to tackle the syntactic heterogeneity in GIS by bringing the GIS data in common format. There are several attempts for GIS interoperability and integration proposed in the literature [5]. A number of proven and well–established methods exist that allow heterogeneous data sources to communicate, including object–oriented approaches, data warehousing and mediators and ontologies.
The most common approach towards GIS interoperability involves the direct translation of geographic data from one vendor or standard file format into another. This is to say, all data should be ported into one common format syntactically. Once the syntax of GIS data is standardized, the sharing of data across organization boundaries will become easier. However, these formats translations can lead to information loss. Also the common data structure should be forth defined, universally acceptable and usually is based on one of existing GIS standards. One important initiative to achieve GIS interoperability is by Open Geospatial Consortium (OGC). This is an association looking to define a set of requirements, standards, and specifications that will support GIS interoperability. But OGC standardized method will not alone suffice to achieve interoperability globally. There are reasons why standardization will not be a complete solution:
- Constructing and maintaining a single, integrated standard data model is difficult problem.
- There will always be a requirement to communicate with information sources that do not conform defined data model standard (legacy systems).
- Existing, legacy information sources have own data models, and there are needs for data conversion from domain model to common model.
2.2 Semantic Heterogeneity and Ontology
Semantic heterogeneity deals primarily in heterogeneous data semantics. Data from one source is useless if other can’t understand it properly. Semantic heterogeneity of the data sources causes serious problems. Since organizations are involved in data creation independently, it is quite unnatural to expect that they will use the same terminology for data properties. Domain experts use the concepts and terminology specific for their respective field of expertise, and use different parameters and different languages to express their model of a concept. Humans use their “common sense”, i.e. their knowledge about the world, to translate the meaning of foreign set of concepts and terms in their own terminology. Although the data sets might be exchangeable (e.g. roads in one agency can mean quite something different for another agency), several questions arises on the semantics of the data. How can the geographic information systems of two agencies with different understanding and models be made interoperable? What are the semantic differences that should be addressed in constructing data sharing environments and developing cross-standard exchange mechanisms? Software systems usually do not have any knowledge about the world and have to explicitly be told how to translate one term into another.
OGC specified base schemas restrict the geometric semantics for encoding GIS data in GML. To share data between communities required that applications must deal not only with geometric and the thematic aspects, but also with the semantic meaning of spatial data. Semantic properties are the definition of the entity or phenomenon. Unlike geometry, which does not vary from one GIS to the next, semantics may vary from group to group in the same way that the conceptual view does. Definitions and meanings may vary slightly or radically, as in the different ways that farmers and civil engineers might define roads.