A Taxonomic Problem

Gilmour, J. S. L.

doi:10.1038/1391040a0

Cited by 66 publications

(17 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The relative frequencies zero and one are simply not in concordance with this fundamental principle, unless we decide, possibly with very little evidence, that certain tests are decisive for membership in a specific class. Gilmour (1937) discussed the distinction between natural and artificial classification and concluded that the more predictions there are that can be made regarding an item on the basis of its membership in a certain class, the more natural the classification is. The maximal predictive classification of Gower(1974) explained in Section 4 formalizes this idea.…”

Section: Discussionmentioning

confidence: 98%

Numerical taxonomy and the principle of maximum entropy

Gyllenberg

Koski

1996

Journal of Classification

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 98%

Numerical taxonomy and the principle of maximum entropy

Gyllenberg

Koski

1996

Journal of Classification

View full text Add to dashboard Cite

show abstract

“…This general stance regarding taxonomy was explicit in the early work of Gilmour (1937), usually cited as one of the philosophical inspirations for the pheneticist view: '…a natural classification can be used for a great variety of purposes, while an artificial one serves only the limited purpose for which it was constructed…The difference between the two is one of degree only' (Gilmour 1937(Gilmour , 1040(Gilmour -1041. Far from being philosophically unusual, this position falls squarely within an influential tradition of theorising about natural kinds: for some purposes we may sort things by electrical conductivity, for some purposes by density, for some purposes by thermal conductivity.…”

Section: Metaphysical Problems: Objective Taxonomymentioning

confidence: 98%

Pheneticism reconsidered

Lewens

2011

Biol Philos

View full text Add to dashboard Cite

The pheneticist philosophy holds that biological taxa are clusters of entities united by a form of all-things-considered resemblance. This view of taxonomy has come in for almost universal criticism from philosophers, and has received little praise from biologists, over the past 30 years or so. This article defends a modest pheneticism, understood as part of a pluralist view of taxonomy. First, phenetic approaches to taxonomy are alive and well in biological practice, especially in the areas of microbiology and botany. Second, the pheneticist notion of overall similarity is defensible, and is implicitly endorsed even by those (such as Quine) usually implicated in attacks on similarity. Third, there are limited biological domains within which pheneticism's conception of species as kinds (rather than heterogeneous individuals) remains applicable.

show abstract

“…Gilmour [21] held the view that the more predictions there are that can be made regarding an item on the basis of its class membership the more natural the classification is. Gower [23,24] made this idea both taxonomically and mathematically precise by constructing for any classification into k classes a list of values (class predictors) that he used to predict the properties of any individual belonging to a given class.…”

Section: Introductionmentioning

confidence: 99%

Classification of Binary Vectors by Stochastic Complexity

Gyllenberg

Koski

Verlaan

1997

Journal of Multivariate Analysis

View full text Add to dashboard Cite

Stochastic complexity is treated as a tool of classification, i.e., of inferring the number of classes, the class descriptions, and the class memberships for a given data set of binary vectors. The stochastic complexity is evaluated with respect to the family of statistical models defined by finite mixtures of multivariate Bernoulli distributions obtained by the principle of maximum entropy. It is shown that stochastic complexity is asymptotically related to the classification maximum likelihood estimate. The formulae for stochastic complexity have an interpretation as minimum code lengths for certain universal source codes for storing the binary data vectors and their assignments into the classes in a classification. There is also a decomposition of the classification uncertainty in a sum of an intraclass uncertainty, an interclass uncertainty, and a special parsimony term. It is shown that minimizing the stochastic complexity amounts to maximizing the information content of the classification. An algorithm of alternating minimization of stochastic complexity is given. We discuss the relation of the method to the AUTOCLASS system of Bayesian classification. The application of classification by stochastic complexity to an extensive data base of strains of Enterobacteriaceae is described.

show abstract

A Taxonomic Problem

Cited by 66 publications

References 0 publications

Numerical taxonomy and the principle of maximum entropy

Numerical taxonomy and the principle of maximum entropy

Pheneticism reconsidered

Classification of Binary Vectors by Stochastic Complexity

Contact Info

Product

Resources

About