The Poisson-Influenced K-Means Algorithm, a Maximum-Likelihood Procedure for Clusters with a Known Probability Distribution

Published

April 25, 2016

Author(s)

Zachary H. Levine, Brian P. Morris

Abstract

We present an implementation of the Poisson-Influenced K-Means Algorithm (PIKA), first developed to characterize the output of a superconducting transition edge sensor (TES) in the few-photon- counting regime. The algorithm seeks to group a number of data into several clusters that minimize their distances from their means, as in classical K-means clustering, but with the added knowledge that the cluster sizes should follow a Poisson distribution.

Citation

The Mathematica Journal

Volume

18

Issue

4

Pub Type

Journals

Download Paper

DOI Link

Keywords

The Poisson-Influenced K-Means Algorithm , Maximum-Likelihood , transition edge sensor , TES

Citation

Levine, Z. and Morris, B. (2016), The Poisson-Influenced K-Means Algorithm, a Maximum-Likelihood Procedure for Clusters with a Known Probability Distribution, The Mathematica Journal, [online], https://doi.org/10.3888/tmj.18-3 (Accessed April 16, 2024)