A deterministic annealing approach to clustering

Rose, Kenneth; Gurewwitz, E.; Fox, Geoffrey

doi:10.1016/0167-8655(90)90010-y

Cited by 343 publications

(180 citation statements)

References 7 publications

Supporting

Mentioning

178

Contrasting

Unclassified

Order By: Relevance

“…Here, we propose a generalization of maximum likelihood for mixture models which is known as annealing and is based on an entropic regularization term. The resulting method is called Tempered Expectation Maximization (TEM) and is closely related to deterministic annealing (Rose, Gurewitz, & Fox, 1990). The combination of deterministic annealing with the EM algorithm has been investigated before in Ueda and Nakano (1998), Hofmann, Puzicha, and Jordan (1999).…”

Section: Model Fitting Revisited: Improving Generalization By Temperementioning

confidence: 99%

Untitled

2001

View full text Add to dashboard Cite

Abstract. This paper presents a novel statistical method for factor analysis of binary and count data which is closely related to a technique known as Latent Semantic Analysis. In contrast to the latter method which stems from linear algebra and performs a Singular Value Decomposition of co-occurrence tables, the proposed technique uses a generative latent class model to perform a probabilistic mixture decomposition. This results in a more principled approach with a solid foundation in statistical inference. More precisely, we propose to make use of a temperature controlled version of the Expectation Maximization algorithm for model fitting, which has shown excellent performance in practice. Probabilistic Latent Semantic Analysis has many applications, most prominently in information retrieval, natural language processing, machine learning from text, and in related areas. The paper presents perplexity results for different types of text and linguistic data collections and discusses an application in automated document indexing. The experiments indicate substantial and consistent improvements of the probabilistic method over standard Latent Semantic Analysis.

show abstract

Section: Model Fitting Revisited: Improving Generalization By Temperementioning

confidence: 99%

Untitled

2001

View full text Add to dashboard Cite

show abstract

“…Miyamoto and Mukaidono [3] considered the singularity in the hard clustering which implies the case where proper partition is not obtained by the Lagrangian multiplier method, and introduced an entropy term as the regularization term with a positive parameter into the objective function of k-Means clustering. Because the fuzzification technique derives the similar algorithm to that of entropy-constrained fuzzy clustering by Deterministic Annealing (DA) [4], the clustering model is often compared with probabilistic mixture models [5]. Then, Ichihashi et al [6] proposed a clustering algorithm, which is similar to the EM algorithm for Gaussian Mixture Models (GMMs), by using the regularization technique with Kullback-Leibler divergences (K-L information).…”

Section: Introductionmentioning

confidence: 99%

Several Formulations for Graded Possibilistic Approach to Fuzzy Clustering

Honda

Ichihashi

Notsu

et al. 2006

Rough Sets and Current Trends in Computing

View full text Add to dashboard Cite

Abstract. Fuzzy clustering is a useful tool for capturing intrinsic structure of data sets. This paper proposes several formulations for soft transition of fuzzy memberships from probabilistic partition to possibilistic one. In the proposed techniques, the free memberships are given by introducing additional penalty term used in Possibilistic c-Means. The new features of the proposed techniques are demonstrated in several numerical experiments.

show abstract

“…This task is performed minimizing a prescribed cost function that is to be adapted to the problem under investigation. In a recent paper [1] it has been shown that a particular clustering algorithm, the so-called Deterministic Annealing (DA) [2,3,4], can be adapted to the study of the hadronic jets in high energy e + e − scattering. Essentially, DA can give the same results of the standard Durham algorithm in a faster way, as a consequence of a lower computational complexity.…”

Section: Introductionmentioning

confidence: 99%

Deterministic annealing as a jet clustering algorithm in hadronic collisions

Angelini

Nardulli

Nitti³

et al. 2004

Physics Letters B

View full text Add to dashboard Cite

We show that a general purpose clusterization algorithm, Deterministic Annealing, can be adapted to the problem of jet identification in particle production by high energy collisions. In particular we consider the problem of jet searching in events generated at hadronic colliders. Deterministic Annealing is able to reproduce the results obtained by traditional jet algorithms and to exhibit a higher degree of flexibility.

show abstract

A deterministic annealing approach to clustering

Cited by 343 publications

References 7 publications

Untitled

Untitled

Several Formulations for Graded Possibilistic Approach to Fuzzy Clustering

Deterministic annealing as a jet clustering algorithm in hadronic collisions

Contact Info

Product

Resources

About