Abstract | ||
---|---|---|
Large repositories of spatial data have been formed in various applications such as Geographic Information Systems (GIS), environmental studies, banking, etc. The increasing demand for knowledge residing inside these databases has attracted much attention to the field of Spatial Data Mining. Due to the common complexity and huge size of spatial databases the aspect of efficiency is of the main concerns in spatial knowledge discovery algorithms. In this paper, we introduce two novel nature-inspired algorithms for efficient discovery of spatial trends, as one of the most valuable patterns in spatial databases. The algorithms are developed using ant colony optimization and evolutionary search. We empirically study and compare the efficiency of the proposed algorithms on a real banking spatial database. The experimental results clearly confirm the improvement in performance and effectiveness of the discovery process compared to the previously proposed methods. |
Year | DOI | Venue |
---|---|---|
2006 | 10.1007/11875581_167 | IDEAL |
Keywords | Field | DocType |
real banking spatial database,spatial knowledge discovery algorithm,spatial data,spatial databases,spatial trend,efficient discovery,mining trend pattern,discovery process,spatial data mining,geographic information systems,proposed algorithm,nature-inspired approach,geographic information system,ant colony optimization,spatial database,knowledge discovery,empirical study | Data field,Data warehouse,Spatial analysis,Geographic information system,Data mining,Computer science,Artificial intelligence,Spatial database,Genetic algorithm,Knowledge extraction,Business process discovery,Database,Machine learning | Conference |
Volume | ISSN | ISBN |
4224 | 0302-9743 | 3-540-45485-3 |
Citations | PageRank | References |
2 | 0.37 | 12 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ashkan Zarnani | 1 | 17 | 3.59 |
Masoud Rahgozar | 2 | 72 | 8.77 |
Caro Lucas | 3 | 1501 | 103.34 |