A Novel Distance Measure for Interval Data

Interval data is attracting attention from the data analysi s community due to its ability to describe complex concepts. Since clust ering is an important data analysis tool, extending these techniques to interval dat is important. Applying traditional clustering methods on interval data los es information inherited in this particular data type. This paper proposes a novel dis sim larity measure which explores the internal structure of intervals in a prob abilistic manner based on domain knowledge. Our experiments show that interval clu stering based on the proposed dissimilarity measure produces meaningful re s lts.