Formal methods in pattern recognition: A review

Nieddu, Luciano; Patrizi, Giacomo

doi:10.1016/s0377-2217(98)00368-3

Cited by 46 publications

(35 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The algorithm stops when set M becomes empty. The convergence of the algorithm in a finite number of steps is proved in a number of ways [10]. The pseudo-code for the T.R.A.C.E algorithm is given in Fig.…”

Section: The Trace Algorithmmentioning

confidence: 99%

“…In fact, this evaluation can help in detecting some errors of the expert (supervisor) that has previously classified the data (recognition with imperfect supervisor). Various techniques can be applied to correct outliers [7,10].…”

Section: Definition 21mentioning

confidence: 99%

“…Some of the latest results have been obtained in medical diagnosis, particularly in the field of human-assisted reproduction [7], and in protein secondary structure classification [8]. Our algorithm is a revised version of the Total Recognition by Adaptive Classification Experiments (T.R.A.C.E) [9,10] algorithm, which partitions the input space into subspaces such that each subspace belongs to one and only one class, and thus the algorithm obtains a complete classification of the training set.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

K-T.R.A.C.E: A kernel -means procedure for classification

Cifarelli

Nieddu

Şeref

et al. 2007

Computers & Operations Research

View full text Add to dashboard Cite

Section: The Trace Algorithmmentioning

confidence: 99%

Section: Definition 21mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

K-T.R.A.C.E: A kernel -means procedure for classification

Cifarelli

Nieddu

Şeref

et al. 2007

Computers & Operations Research

View full text Add to dashboard Cite

“…Several applications have been reported in the literature to measure the insulin sensitivity index for dierent patient populations, including diabetes [71,111], aging [25,73], hyperthyroidism [83], hyperparathyroidism [55], myotonic dystrophy [85], pregnancy [19], gynaecological conditions [21], obesity [13], hypertension [19], cirrhosis [58,64], ethnical subpopulation [42,43,50,69,98], siblings of diabetic patients, [29,51,88] and during pharmacological tests [62].…”

Section: Glucose±insulin Kineticsmentioning

confidence: 99%

Operational research techniques in medical treatment and diagnosis: A review

Bartolozzi

Gaetano

Lena

et al. 2000

European Journal of Operational Research

View full text Add to dashboard Cite

Many modern techniques for the diagnosis of pathological states in humans and for their subsequent treatment can be posed as nonlinear identi®cation problems of essentially nonlinear dynamic systems or as nonlinear optimal control problems. It can be shown that the linearised versions of such models are inadequate and do not represent at all well the complexity of the problem. Thus, nonlinear estimation and control techniques are required for progress to be made in this ®eld.The aim of this review is to examine some models suggested in the medical literature for the modelling of certain medical treatments and diagnoses. Then examine how these models can be enriched by using Operational Research techniques so that a better control is provided on the diagnosis and the treatment, as well as the formulation of more precise models of the phenomenon.The review will present some applications both therapeutic and diagnostic that have appeared in the literature. Special interest will be bestowed on hyperthermic systems in oncological treatment and glucose±insulin dynamics for diabetic patients, while heart dynamics and magnetic resonance imaging will also receive attention. These applications are good examples to show the advantages of Operational Research methods in this ®eld of endeavour.The outline of the paper is the following. After the introduction, in section two a brief description of nonlinear system models of phenomena will be given, for de®nitional and descriptive purposes. In section three a discussion of how to apply System theory in the medical ®eld will be presented, together with an analysis of the possible bene®ts. In section four some applications of dynamical models to medical diagnosis and treatment will be described, while in section ®ve the appropriate conclusions will be stated. Ó 2000 Published by Elsevier Science B.V. All rights reserved. 0377-2217/00/$ -see front matter Ó 2000 Published by Elsevier Science B.V. All rights reserved. PII: S 0 3 7 7 -2 2 1 7 ( 9 9 ) 0 0 0 1 7 -X

show abstract

“…[21], which generalize both prototype-based and exemplarbased classi ers, can be used to discover the \typicality structure" of the learnt category, detecting how the class is decomposable in some subclasses and their typicality grade within the class [22].…”

Section: Introductionmentioning

confidence: 99%

A prototype-based classification using extended imperialist competitive algorithm

et al. 2017

View full text Add to dashboard Cite

KEYWORDSPrototype-based classi cation; Imperialist competitive algorithm; UCI machine learning repository.Abstract. Classi cation is an important machine learning technique used to predict group membership for data instances. In this paper, we propose an e cient prototypebased classi cation approach in the data classi cation literature by a novel soft-computing approach based on extended imperialist competitive algorithm. The novel classi er is called EICA. The goal is to determine the best places of the prototypes. EICA is evaluated under three di erent tness functions on twelve typical test datasets from the UCI Machine Learning Repository. The performance of the proposed EICA is compared with well-developed algorithms in classi cation including original Imperialist Competitive Algorithm (ICA), the Arti cial Bee Colony (ABC), the Fire y Algorithm (FA), the Particle Swarm Optimization (PSO), the Gravitational Search Algorithm (GSA), the Grouping Gravitational Search Algorithm (GGSA), and nine well-known classi cation techniques in the literature. The analysis results show that EICA provides encouraging results in contrast to other algorithms and classi cation techniques.

show abstract

Formal methods in pattern recognition: A review

Cited by 46 publications

References 45 publications

K-T.R.A.C.E: A kernel -means procedure for classification

K-T.R.A.C.E: A kernel -means procedure for classification

Operational research techniques in medical treatment and diagnosis: A review

A prototype-based classification using extended imperialist competitive algorithm

Contact Info

Product

Resources

About