Geo-Spatial Clustering with User-Specified Constraints

Capturing application semantics and allowing a human analyst to express his focus in mining have been the motivation for several recent studies on constrained mining. In this paper, we introduce and study the problem of constrained clustering| nding clusters that satisfy certain user-speci ed constraints. We argue that this problem arises naturally in practice. Two types of constraints are discussed in this paper. The rst type of constraints are imposed by physical obstacles that exist in the region of clustering. The second type of constraints are SQL constraints which every cluster must satisfy. We provide a prelimary introduction to both types of constraints and discuss some techniques for