Showing 1–1 of 1 results for author: Rafsky, L C

Exact clustering in linear time

Authors: Jonathan A. Marshall, Lawrence C. Rafsky

Abstract: The time complexity of data clustering has been viewed as fundamentally quadratic, slowing with the number of data items, as each item is compared for similarity to preceding items. Clustering of large data sets has been infeasible without resorting to probabilistic methods or to capping the number of clusters. Here we introduce MIMOSA, a novel class of algorithms which achieve linear time computa… ▽ More The time complexity of data clustering has been viewed as fundamentally quadratic, slowing with the number of data items, as each item is compared for similarity to preceding items. Clustering of large data sets has been infeasible without resorting to probabilistic methods or to capping the number of clusters. Here we introduce MIMOSA, a novel class of algorithms which achieve linear time computational complexity on clustering tasks. MIMOSA algorithms mark and match partial-signature keys in a hash table to obtain exact, error-free cluster retrieval. Benchmark measurements, on clustering a data set of 10,000,000 news articles by news topic, found that a MIMOSA implementation finished more than four orders of magnitude faster than a standard centroid implementation. △ Less

Submitted 27 February, 2017; v1 submitted 17 February, 2017; originally announced February 2017.

Comments: 16 pages, 4 figures

ACM Class: H.3.3; I.5.3; F.2.2; E.1