Algorithms of maximum likelihood data clustering with applications

DOI10.1016/S0378-4371(02)00974-3zbMATH Open1001.65013arXivcond-mat/0204202MaRDI QIDQ1848191FDOQ1848191

Authors: Lorenzo Giada, Matteo Marsili 

Publication date: 19 November 2002

Published in: Physica A (Search for Journal in Brave)

Abstract: We address the problem of data clustering by introducing an unsupervised, parameter free approach based on maximum likelihood principle. Starting from the observation that data sets belonging to the same cluster share a common information, we construct an expression for the likelihood of any possible cluster structure. The likelihood in turn depends only on the Pearson's coefficient of the data. We discuss clustering algorithms that provide a fast and reliable approximation to maximum likelihood configurations. Compared to standard clustering methods, our approach has the advantages that i) it is parameter free, ii) the number of clusters need not be fixed in advance and iii) the interpretation of the results is transparent. In order to test our approach and compare it with standard clustering algorithms, we analyze two very different data sets: Time series of financial market returns and gene expression data. We find that different maximization algorithms produce similar cluster structures whereas the outcome of standard algorithms has a much wider variability.

Full work available at URL: https://arxiv.org/abs/cond-mat/0204202

Recommendations

• Fuzzy clustering algorithms based on the maximum likelihood principle
• A new approach to clustering
• Information-based clustering
• Clustering by maximizing a fuzzy classification maximum likelihood criterion
• How Many Clusters? An Information-Theoretic Perspective

zbMATH Keywords

time seriesPearson coefficient

Mathematics Subject Classification ID

Numerical smoothing, curve fitting (65D10)

Cites Work

• Title not available (Why is that?)
• Optimization by simulated annealing
• Equation of state calculations by fast computing machines
• The EM algorithm for graphical association models with missing data

Cited In (12)

• Power mapping with dynamical adjustment for improved portfolio optimization
• SPANNING TREES AND BOOTSTRAP RELIABILITY ESTIMATION IN CORRELATION-BASED NETWORKS
• Financial time series analysis using Total-CApEn and Avg-CApEn with cumulative histogram matrix
• A review of two decades of correlations, hierarchies, networks and clustering in financial markets
• Agglomerative likelihood clustering
• Dissecting financial markets: sectors and states
• Scale invariance and criticality in financial markets
• Criticality of mostly informative samples: a Bayesian model selection approach
• Experiments with, and on, algorithms for maximum likelihood clustering
• Multiscale adaptive multifractal analysis and its applications
• Identifying dominant industrial sectors in market states of the S&P 500 financial data
• On sampling and modeling complex systems

This page was built for publication: Algorithms of maximum likelihood data clustering with applications

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1848191)