Abstract
BIRCH clustering is a widely known approach for clustering, that has influenced much subsequent research and commercial products. The key contribution of BIRCH is the Clustering Feature tree (CF-Tree), which is a compressed representation of the input data. As new data arrives, the tree is eventually rebuilt to increase the compression. Afterward, the leaves of the tree are used for clustering. Because of the data compression, this method is very scalable. The idea has been adopted for example for k-means, data stream, and density-based clustering.
Clustering features used by BIRCH are simple summary statistics that can easily be updated with new data: the number of points, the linear sums, and the sum of squared values. Unfortunately, how the sum of squares is then used in BIRCH is prone to catastrophic cancellation.
We introduce a replacement cluster feature that does not have this numeric problem, that is not much more expensive to maintain, and which makes many computations simpler and hence more efficient. These cluster features can also easily be used in other work derived from BIRCH, such as algorithms for streaming data. In the experiments, we demonstrate the numerical problem and compare the performance of the original algorithm compared to the improved cluster features.
Part of the work on this paper has been supported by Deutsche Forschungsgemeinschaft (DFG) within the Collaborative Research Center SFB 876 “Providing Information by Resource-Constrained Analysis”, project A2. https://sfb876.tu-dortmund.de/.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Aggarwal, C.C., Han, J., Wang, J., Yu, P.S.: A framework for clustering evolving data streams. In: VLDB, pp. 81–92 (2003). https://doi.org/10.1016/B978-012722442-8/50016-1
Aggarwal, C.C., Han, J., Wang, J., Yu, P.S.: A framework for projected clustering of high dimensional data streams. In: VLDB, pp. 852–863 (2004). https://doi.org/10.1016/B978-012088469-8.50075-9
Ankerst, M., Breunig, M.M., Kriegel, H., Sander, J.: OPTICS: ordering points to identify the clustering structure. In: SIGMOD, pp. 49–60 (1999). https://doi.org/10.1145/304182.304187
Arthur, D., Vassilvitskii, S.: k-means++: the advantages of careful seeding. In: SODA, pp. 1027–1035 (2007)
Bonner, R.E.: On some clustering techniques. IBM J. Res. Dev. 8(1), 22–32 (1964)
Bradley, P.S., Fayyad, U.M., Reina, C.: Scaling clustering algorithms to large databases. In: KDD, pp. 9–15 (1998)
Breunig, M.M., Kriegel, H., Kröger, P., Sander, J.: Data bubbles: quality preserving performance boosting for hierarchical clustering. In: SIGMOD, pp. 79–90 (2001). https://doi.org/10.1145/375663.375672
Breunig, M.M., Kriegel, H.-P., Sander, J.: Fast hierarchical clustering based on compressed data and OPTICS. In: Zighed, D.A., Komorowski, J., Żytkow, J. (eds.) PKDD 2000. LNCS (LNAI), vol. 1910, pp. 232–242. Springer, Heidelberg (2000). https://doi.org/10.1007/3-540-45372-5_23
Cao, F., Ester, M., Qian, W., Zhou, A.: Density-based clustering over an evolving data stream with noise. In: SDM, pp. 328–339 (2006). https://doi.org/10.1137/1.9781611972764.29
Chiu, T., Fang, D., Chen, J., Wang, Y., Jeris, C.: A robust and scalable clustering algorithm for mixed type attributes in large database environment. In: KDD, pp. 263–268 (2001). https://doi.org/10.1145/502512.502549
Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the EM algorithm. J. Royal Stat. Soc. Series B 39(1), 1–38 (1977)
Estivill-Castro, V.: Why so many clustering algorithms: a position paper. SIGKDD Explor. 4(1), 65–75 (2002)
Fichtenberger, H., Gillé, M., Schmidt, M., Schwiegelshohn, C., Sohler, C.: BICO: BIRCH meets coresets for k-means clustering. In: Bodlaender, H.L., Italiano, G.F. (eds.) ESA 2013. LNCS, vol. 8125, pp. 481–492. Springer, Heidelberg (2013). https://doi.org/10.1007/978-3-642-40450-4_41
Fraley, C., Raftery, A.E.: Bayesian regularization for normal mixture estimation and model-based clustering. J. Classif. 24(2), 155–181 (2007). https://doi.org/10.1007/s00357-007-0004-5
Ganti, V., Gehrke, J., Ramakrishnan, R.: DEMON: mining and monitoring evolving data. IEEE Trans. Knowl. Data Eng. 13(1), 50–63 (2001). https://doi.org/10.1109/69.908980
Han, J., Kamber, M., Pei, J.: Data Mining: Concepts and Techniques, 3rd edn. Morgan Kaufmann, Burlington (2011)
Kranen, P., Assent, I., Baldauf, C., Seidl, T.: The ClusTree: indexing micro-clusters for anytime stream mining. Knowl. Inf. Syst. 29(2), 249–272 (2011). https://doi.org/10.1007/s10115-010-0342-8
Kriegel, H.-P., Schubert, E., Zimek, A.: The (black) art of runtime evaluation: are we comparing algorithms or implementations? Knowl. Inf. Syst. 52(2), 341–378 (2017). https://doi.org/10.1007/s10115-016-1004-2
Lang, A., Schubert, E.: BETULA: numerically stable CF-trees for BIRCH clustering. CoRR abs/2006.12881 (2020). https://arxiv.org/abs/2006.12881
Ng, R.T., Han, J.: CLARANS: a method for clustering objects for spatial data mining. IEEE Trans. Knowl. Data Eng. 14(5), 1003–1016 (2002). https://doi.org/10.1109/TKDE.2002.1033770
Schubert, E., Gertz, M.: Numerically stable parallel computation of (co-)variance. In: SSDBM, pp. 10:1–10:12 (2018). https://doi.org/10.1145/3221269.3223036
Schubert, E., Zimek, A.: ELKI: a large open-source library for data analysis - ELKI release 0.7.5 “Heidelberg”. CoRR abs/1902.03616 (2019)
Zhang, T.: Data clustering for very large datasets plus applications. Technical report 1355, University of Wisconsin Madison (1996)
Zhang, T., Ramakrishnan, R., Livny, M.: BIRCH: an efficient data clustering method for very large databases. In: SIGMOD, pp. 103–114 (1996). https://doi.org/10.1145/233269.233324
Zhang, T., Ramakrishnan, R., Livny, M.: BIRCH: a new data clustering algorithm and its applications. Data Min. Knowl. Discov. 1(2), 141–182 (1997). https://doi.org/10.1023/A:1009783824328
Zhang, T., Ramakrishnan, R., Livny, M.: Fast density estimation using CF-kernel for very large databases. In: KDD, pp. 312–316 (1999). https://doi.org/10.1145/312129.312266
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Lang, A., Schubert, E. (2020). BETULA: Numerically Stable CF-Trees for BIRCH Clustering. In: Satoh, S., et al. Similarity Search and Applications. SISAP 2020. Lecture Notes in Computer Science(), vol 12440. Springer, Cham. https://doi.org/10.1007/978-3-030-60936-8_22
Download citation
DOI: https://doi.org/10.1007/978-3-030-60936-8_22
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-60935-1
Online ISBN: 978-3-030-60936-8
eBook Packages: Computer ScienceComputer Science (R0)