Density-Based Data Analysis and Similarity Search

Abstract

Similarity search in database systems is becoming an increasingly important task in modern application domains such as multimedia, molecular biology, medical imaging, computer aided engineering, marketing and purchasing assistance as well as many others. Furthermore, the feature transformations and distance measures used in similarity search build the foundation of sophisticated data analysis and mining techniques. In this chapter, we show how visualizing cluster hierarchies describing a database of objects can aid the user in the time consuming task to find similar objects and discover interesting patterns. We present related work and explain its shortcomings which led to the development of our new methods. Based on reachability plots, we introduce methods for visually exploring a data set in multiple representations and comparing multiple similarity models. Furthermore, we present a new method for automatically extracting cluster hierarchies from a given reachability plot which allows a user to browse the database for similarity search. We integrated our new method in a prototype which serves two purposes, namely visual data analysis and a new way of object retrieval called navigational similarity search.

Extracted Key Phrases

13 Figures and Tables

Cite this paper

@inproceedings{Kriegel2006DensityBasedDA, title={Density-Based Data Analysis and Similarity Search}, author={Hans-Peter Kriegel and Stefan Brecheisen and Peer Kr{\"{o}ger and Martin Pfeifle and Matthias Schubert and Arthur Zimek}, year={2006} }