Universiti Teknologi Malaysia Institutional Repository

Spatio-temporal object relationships in hydrological data: a research perspective

Awan, A. Majid and Md Sap, Mohd Noor (2004) Spatio-temporal object relationships in hydrological data: a research perspective. Jurnal Teknologi Maklumat, 16 (1). pp. 76-89. ISSN 0128-3790

[img] PDF
528Kb

Abstract

The objects in geographical information systems are represented in a context extremely rich of relationships among them. Complexity of geographical data precludes the use of general purpose relationship discovery and data analysis techniques. Most of the decision support systems in the geographic domain depend heavily on the effective and efficient processing of neighbourhood relations and cause and consequence type relations. Three important techniques, which try to exploit these object relationships for capturing meaningful patterns, are: clustering, classification and association analysis. This paper discusses the importance of exploiting spatio-temporal object relationship, the three important techniques which try to exploit these relationships for effective and efficient data analysis, identifies a few of the related research issues which need to be addressed, and then proposes a few solutions to the issues. For instance clustering high dimensional data is challenging when the clusters are of widely differing shapes, sizes, and densities, and when the data contains noise and outliers. Research in clustering spatial data has mostly focused on effectiveness and scalability for large databases. However, only a very few of them have taken into account constraints that may be present in the data or constraints on the clustering. These constraints have significant influence on the results of the clustering process of large spatial data. For Association analysis, there are many problems for deriving the spatial and spatio-temporal relationships hidden in the data, such as: encoding spatial relationships for all combinations of geographic objects, data integration, data classification, the representation and calculation of spatial relationships, and strategies for finding interesting rules. It is necessary to address these issues for effective and efficient association analysis of the data. For spatial classification step, as the data inputs for spatial analysis have both spatial and non-spatial features, so it is required to consider both type of features including auto-correlation for mapping the data in meaningful classes. Finally, as a case study, the application of these techniques exploiting spatio-temporal object relationships in the context of developing a decision support system in hydrological environment is briefly described.

Item Type:Article
Uncontrolled Keywords:spatial data, spatio-temporal data, cluster analysis, classification, association rule
Subjects:Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:Computer Science and Information System (Formerly known)
ID Code:3422
Deposited By: Mrs Rozilawati Dollah @ Md Zain
Deposited On:24 May 2007 03:57
Last Modified:30 May 2011 03:09

Repository Staff Only: item control page