Comparing two density-based clustering methods for reducing very large spatio-temporal dataset

摘要

Cluster-based mining methods have proven to be a successful method for the reduction of very large spatio-temporal datasets. These datasets are often very large and difficult to analyse. Clustering methods can be used to decrease the large size of original data by retrieving its useful knowledge as representatives. As a consequence, instead of dealing with a large size of raw data, we can use these representatives to visualise or to analyse without losing important information. In this paper, we compare our two clustering-based approaches for reducing large spatio-temporal datasets. Both approaches are based on the combination of density-based and graph-based clustering. The first one takes into account the Shared Nearest Neighbour degree and the second one applies the Euclidean metric distance radius to determine the nearest neighbour similarity. We also present and discuss preliminary results for this comparison.