Preserving consensus hierarchies

In numerical taxonomy we often have the task of finding a consensus hierarchy for a given set of hierarchies. This consensus hierarchy should reflect the substructures which are common to all hierarchies of the set. Because there are several kinds of substructures in a hierarchy, the general axiom to preserve common substructures leads to different axioms for each kind of substructure. In this paper we consider the three substructurescluster, separation, andnesting, and we give several characterizations of hierarchies preserving these substructures. These characterizations facilitate interpretation of axioms for preserving substructures and the examination of properties of consensus methods. Finally some extensions concerning the preserving of qualified substructures are discussed.The author is grateful to the editor and the referees for their helpful suggestions and to H. J. Bandelt for his comments on an earlier version of this paper.     

On the use of ordered sets in problems of comparison and consensus of classifications

Ordered set theory provides efficient tools for the problems of comparison and consensus of classifications Here, an overview of results obtained by the ordinal approach is presented Latticial or semilatticial structures of the main sets of classification models are described Many results on partitions are adaptable to dendrograms; many results on n-trees hold in any median semilattice and thus have counterparts on ordered trees and Buneman (phylogenetic) trees For the comparison of classifications, the semimodularity of the ordinal structures involved yields computable least-move metrics based on weighted or unweighted elementary transformations In the unweighted case, these metrics have simple characteristic properties For the consensus of classifications, the constructive, axiomatic, and optimization approaches are considered Natural consensus rules (majoritary, oligarchic, ) have adequate ordinal formalizations A unified presentation of Arrow-like characterization results is given In the cases of n-trees, ordered trees and Buneman trees, the majority rule is a significant example where the three approaches convergeThe authors would like to thank the anonymous referees for helpful suggestions on the first draft of this paper, and W H E Day for his comments and his significant improvements of style     

Comparison of classifications using measures intermediate between metric dissimilarity and consensus similarity

Two fundamental approaches to the comparison of classifications (e g, partitions on the same finite set of objects) can be distinguished One approach is based upon measures of metric dissimilarity while the other is based upon measures of similarity, or consensus These approaches are not necessarily simple complements of each other Instead, each captures different, limited views of comparison of two classifications The properties of these measures are clarified by their relationships to Day's complexity models and to association measures of numerical taxonomy The two approaches to comparison are equated with the use of separation and minimum value sensitive measures, suggesting the potential application of an intermediate sensitive measure to the problem of comparison of classifications Such a measure is a linear combination of separation sensitive and minimum value sensitive components The application of these intermediate measures is contrasted with the two extremes The intermediate measure for the comparison of classifications is applied to a problem of character weighting arising in the analysis of Australian stream basinsWe thank Bill Day, Mike Austin, Peter Minchin and two anonymous referees for many helpful comments We also thank P Arabie for useful discussion of consensus methods and character weighting     

One-Mode Classification of a Three-Way Data Matrix

X is the automatic hierarchical classification of one mode (units or variables or occasions) of X on the basis of the other two. In this paper the case of OMC of units according to variables and occasions is discussed. OMC is the synthesis of a set of hierarchical classifications Delta obtained from X; e.g., the OMC of units is the consensus (synthesis) among the set of dendograms individually defined by clustering units on the basis of variables, separately for each given occasion of X. However, because Delta is often formed by a large number of classifications, it may be unrealistic that a single synthesis is representative of the entire set. In this case, subsets of similar (homegeneous) dendograms may be found in Delta so that a consensus representative of each subset may be identified. This paper proposes, PARtition and Least Squares Consensus cLassifications Analysis (PARLSCLA) of a set of r hierarchical classifications Delta. PARLSCLA identifies the best least-squares partition of Delta into m (1 <= m <= r) subsets of homogeneous dendograms and simultaneously detects the closest consensus classification (a median classification called Least Squares Consensus Dendogram (LSCD) for each subset. PARLSCLA is a generalization of the problem to find a least-squares consensus dendogram for Delta. PARLSCLA is formalized as a mixed-integer programming problem and solved with an iterative, two-step algorithm. The method proposed is applied to an empirical data set.     

Character and OTU stability in five taxonomic groups

The character and OTU stability of classifications based on UPGMA clustering and maximum parsimony (MP) trees were compared for 5 datasets (families of angiosperms, families of orthopteroid insects, species of the fish genusIctalurus, genera of the salamander family Salamandridae, and genera of the frog family Myobatrachidae). Stability was investigated by taking different sized random subsamples of OTUs or characters, computing UPGMA clusters and an MP tree, and then comparing the resulting trees with those based on the entire dataset. Agreement was measured by two consensus indices, that of Colless, computed from strict consensus trees, and Stinebrickner's 0.5-consensus index. Tests of character stability generally showed a monotone decrease in agreement with the standard as smaller sets of characters are considered. The relative success of the two methods depended upon the dataset. Tests of OTU stability showed a monotone decrease in agreement for UPGMA as smaller sets of OTUs are considered. But for MP, agreement decreased and then increased again on the same scale. The apparent superiority of UPGMA relative to MP with respect to OTU stability depended upon the dataset. Considerations other than stability, such as computer efficiency or accuracy, will also determine the method of choice for classifications.     

On lattice consensus methods

We investigate the consensus problem for classifications of three types: partitions, dendrograms, and n-trees For partitions or dendrograms, lattice polynomials define natural consensus functions We extend these lattice methods to n-trees, introducing a general class of consensus functions that includes the intersection consensus functions in current use These lattice consensus methods have a number of desirable mathematical properties We prove that they all satisfy the Pareto Axiom For each of the three classification types, we determine which lattice consensus functions satisfy the Betweenness AxiomAuthor partially supported by a research grant from the Faculty Research Committee, Bowling State University     

Structural Classification Analysis of Three-Way Dissimilarity Data

The paper presents a methodology for classifying three-way dissimilarity data, which are reconstructed by a small number of consensus classifications of the objects each defined by a sum of two order constrained distance matrices, so as to identify both a partition and an indexed hierarchy. Specifically, the dissimilarity matrices are partitioned in homogeneous classes and, within each class, a partition and an indexed hierarchy are simultaneously fitted. The model proposed is mathematically formalized as a constrained mixed-integer quadratic problem to be fitted in the least-squares sense and an alternating least-squares algorithm is proposed which is computationally efficient. Two applications of the methodology are also described together with an extensive simulation to investigate the performance of the algorithm.     

Obtaining common pruned trees

Given two or more dendrograms (rooted tree diagrams) based on the same set of objects, ways are presented of defining and obtaining common pruned trees. Bounds on the size of a largest common pruned tree are introduced, as is a categorization of objects according to whether they belong to all, some, or no largest common pruned trees. Also described is a procedure for regrafting pruned branches, yielding trees for which one can assess the reliability of the depicted relationships. The tree obtained by regrafting branches on to a largest common pruned tree is shown to contain all the classes present in the strict consensus tree. The theory is illustrated by application to two classifications of a set of forty-nine stratigraphical pollen spectra.This work was supported by the Science and Engineering Research Council. The authors are grateful to the referees for constructive criticisms of an earlier version of the paper, and to Dr. J.T. Henderson for advice on PASCAL.     

s-Consensus index method: An additional axiom

A consensus index method is an ordered pair consisting of a consensus method and a consensus index Day and McMorris (1985) have specified two minimal axioms, one which should be satisfied by the consensus method and the other by the consensus index The axiom for consensus indices is not satisfied by the s-consensus index In this paper, an additional axiom, which states that a consensus index equal to one implies profile unanimity, is proposed The s-consensus method together with a modification of the s-consensus index (i e, normalized by the number of distinct nontrivial clusters in the profile) is shown to satisfy the two axioms proposed by Day and McMorris and the new axiom     

Evolutionary and psychological effects in pre-evolutionary classifications

Relative frequency of genera as a function of number of species per genus is plotted for six eighteenth-century classifications: Linnaeus' classifications of animals, plants, minerals, and diseases, and Sauvages' classifications of plants and diseases. The distributions for animals and plants form positively skewed hollow curves similar but not identical to those found in modern biological classifications and predicted by mathematical models of evolution. The distributions for minerals and diseases, however, are more nearly symmetric and convex. The difference between the eighteenth-century and modern classifications of animals and plants probably reflects psychological properties of the taxonomists' judgments; but the difference between the classifications of animals and plants and those of minerals and diseases reflects evolutionary properties of the materials classified, since all six classifications were constructed by the same taxonomists using the same methods. Consequently, the observable effects of evolution are strong enough to be detected in classifications constructed before the acceptance of evolutionary theory; and traditional classifications can contain substantial scientific information despite their reliance on incompletely understood processes of judgment.I thank Mae Ling Hum for assistance in data collection, and Dennis G. Fisher, David M. Raup, Thomas D. Wickens, and J. Arthur Woodward for helpful comments on earlier versions of the paper. Computer time was provided by the UCLA Office of Academic Computing.     

The complexity of reconstructing trees from qualitative characters and subtrees

In taxonomy and other branches of classification it is useful to know when tree-like classifications on overlapping sets of labels can be consistently combined into a parent tree. This paper considers the computation complexity of this problem. Recognizing when a consistent parent tree exists is shown to be intractable (NP-complete) for sets of unrooted trees, even when each tree in the set classifies just four labels. Consequently determining the compatibility of qualitative characters and partial binary characters is, in general, also NP-complete. However for sets of rooted trees an algorithm is described which constructs the “strict consensus tree” of all consistent parent trees (when they exist) in polynomial time. The related question of recognizing when a set of subtrees uniquely defines a parent tree is also considered, and a simple necessary and sufficient condition is described for rooted trees. This work was supproted by the Alexander von Humoldt-Stiftung. I wish to thank Andreas Dress, Hans-Jürgen Bandelt and the referees for their helpful comments.     

非遗术语翻译的知识理据创新及其应用研究——以剪纸为例

非遗术语翻译是跨语表达非遗知识的重要实践形式。当前,国内以非遗概念为本位的翻译理据虽然积极促进了非遗译名的规范化实践,但就如何厘清非遗术语核心概念要素这一问题,还缺乏深入的路径反思。文章在评析概念理据的基础上,进一步提出非遗术语翻译的知识理据,即非遗知识具有层级性的特征,可从物质层、行为层和精神层予以划分,且隶属于不同知识层次的术语具有差异性概念特征。文章以剪纸术语系统及其翻译实践为例,论证了知识理据的应用有效性,即知识理据可指导译者在非遗知识体系的场域中明确对象术语的概念内涵与相应的翻译需求。     

Kappa Coefficients for Circular Classifications

Circular classifications are classification scales with categories that exhibit a certain periodicity. Since linear scales have endpoints, the standard weighted kappas used for linear scales are not appropriate for analyzing agreement between two circular classifications. A family of kappa coefficients for circular classifications is defined. The kappas differ only in one parameter. It is studied how the circular kappas are related and if the values of the circular kappas depend on the number of categories. It turns out that the values of the circular kappas can be strictly ordered in precisely two ways. The orderings suggest that the circular kappas are measuring the same thing, but to a different extent. If one accepts the use of magnitude guidelines, it is recommended to use stricter criteria for circular kappas that tend to produce higher values.     

The Haar Wavelet Transform of a Dendrogram

We describe a new wavelet transform, for use on hierarchies or binary rooted trees. The theoretical framework of this approach to data analysis is described. Case studies are used to further exemplify this approach. A first set of application studies deals with data array smoothing, or filtering. A second set of application studies relates to hierarchical tree condensation. Finally, a third study explores the wavelet decomposition, and the reproducibility of data sets such as text, including a new perspective on the generation or computability of such data objects.     

Statistical properties of large published classifications

Large published classifications typically consist of sets (called taxa) hierarchically arranged according to taxonomic rank. A statistical survey of 23 such classification reveals the following distinctive properties. The pattern of mandatory and optional taxonomic ranks is similar to a Guttman scale. Mean taxon size (defined as the number of next-lower-rank taxa per higher-rank taxon) is a U-shaped function of mandatory rank, and averages about seven across ranks with no significant differences between classifications. The variability of taxon size is a decreasing function of mandatory rank. The generality of these properties across classifications suggests that they are determined by the psychology of the classification process. In contrast, there are significant differences between classifications in the variability of taxon size and in the prevalence of optional ranks, both of which are greater in biological than in nonbiological classifications. These differences may reflect the nature of the materials classified. This research was supported by a research grant from the UCLA Academic Senate and by computer time from the UCLA Office of Academic Computing.     

TreeOfTrees Method to Evaluate the Congruence Between Gene Trees

A new method, TreeOfTrees, is proposed to compare X-tree structures obtained from several sets of aligned gene sequences of the same taxa. Its aim is to detect genes or sets of genes having different evolutionary histories. The comparison between sets of trees is based on several tree metrics, leading to a unique tree labelled by the gene trees. The robustness values of its edges are estimated by bootstrapping and consensus procedures that allow detecting subsets of genes having differently evolved. Simulations are performed under various evolutionary conditions to test the efficiency of the method and an application on real data is described. Tests of arboricity and various consensus algorithms are also discussed. A corresponding software package is available.     

A Probability for Classification Based on the Dirichlet Process Mixture Model

In this paper we provide an explicit probability distribution for classification purposes when observations are viewed on the real line and classifications are to be based on numerical orderings. The classification model is derived from a Bayesian nonparametric mixture of Dirichlet process model; with some modifications. The resulting approach then more closely resembles a classical hierarchical grouping rule in that it depends on sums of squares of neighboring values. The proposed probability model for classification relies on a numerical procedure based on a reversible Markov chain Monte Carlo (MCMC) algorithm for determining the probabilities. Some numerical illustrations comparing with alternative ideas for classification are provided.     

Scientific Consensus and Expert Testimony in Courts: Lessons from the Bendectin Litigation

A consensus in a scientific community is often used as a resource for making informed public-policy decisions and deciding between rival expert testimonies in legal trials. This paper contains a social-epistemic analysis of the high-profile Bendectin drug controversy, which was decided in the courtroom inter alia by deference to a scientific consensus about the safety of Bendectin. Drawing on my previously developed account of knowledge-based consensus, I argue that the consensus in this case was not knowledge based, hence courts’ deference to it was not epistemically justified. I draw sceptical lessons from this analysis regarding the value of scientific consensus as a desirable and reliable means of resolving scientific controversies in public life.     

俄语PPP专业术语特点及翻译方略刍议

文章先对俄语PPP专业术语的特点进行了分类总结,然后在所提出的“CLAST”术语翻译原则的基础之上,重点对有俄罗斯本土特色的PPP专业术语的翻译策略和方法提出了相应的建议和意见。     

N-trees as nestings: Complexity,similarity, and consensus

Interpreting a taxonomic tree as a set of objects leads to natural measures of complexity and similarity, and sets natural lower bounds on a consensus tree Interpretations differing as to the kind of objects constituting a tree lead to different measures and consensus Subset nesting is preferred over the clusters (strict consensus) and even the triads interpretations because of its superior expression of shared structure Algorithms for computing the complexity and similarity of trees, as well as a consensus index onto [0,1], are presented for this interpretation The full consensus is defined as the only tree which includes all the nestings shared in a profile of rival trees and whose clusters reflect only nestings shared in the profile The full consensus is proved to exist uniquely for each profile, and to equal the Adams consensusThe author is grateful for the many helpful comments on presentation from Frances McA Adams, William H E Day, and Christopher A Meacham