Data mining in large data sets often requires a sampling or summarization step to form an in-core representation of the data that can be processed more efficiently. Uniform random sampling is frequently used in practice and also frequently criticized because it will miss small clusters. Many natural phenomena are known to follow Zipf's distribution and the inability of uniform sampling to find small clusters is of practical concern. Density Biased Sampling is proposed to probabilistically under-sample dense regions and over-sample light regions. A weighted sample is used to preserve the densities of the original data. Density biased sampling naturally includes uniform sampling as a special case. A memory efficient algorithm is proposed that approximates density biased sampling using only a single scan of the data. We empirically evaluate density biased sampling using synthetic data sets that exhibit varying cluster size distributions. Our proposed method scales linearly and out performs uniform samples when clustering realistic data sets.

23 pages

Return to: SCS Technical Report Collection
School of Computer Science homepage

This page maintained by reports@cs.cmu.edu