|
Modify search criterions
Results n° 1 to 8 of 71 matches
| Title |
Computation of overlapping classes in a graph: application to protein-protein interactions networks |
| Author |
DENOEUD-BELGACEM Lucile |
| Keywords |
Classification by density, Graph, k-means, Mapping, Overlapping classification, Protein-protein interactions networks |
| Topics |
Biology, Classification - Clustering - Partitioning, Graphs, Mathematic models |
| Abstract |
This article describes a method of overlapping classification, in order to compute zones which are dense in edges in a graph. More precisely, the aim is to compute subgraphs in which the density of edges is large compared to the edge-density of the whole graph. These subgraphs may share common vertices. This method is applied to a problem arising in biology: the annotation of proteins. The graphs then represent the observed interactions between proteins. Thanks to the biological principle that proteins involved in the same cellular function interact, the subgraphs provided when the method is applied to the protein-protein interactions networks provide information about the functions of proteins belonging to these subgraphs. This provides a computer-aided tool for the prediction of unknown functions of some proteins. The overlapping allowed by the method depicted here makes it possible to take into account the fact that each protein may be involved into several cellular functions. |
| Number |
187, Fall 2009, special issue: 2007 Meeting of the French-speaking Society of Classification |
| Language |
 French | Read the article
| Title |
Mayer's fitting method and its links to clustering methods |
| Author |
FALGUEROLLES Antoine |
| Keywords |
Classification, History of statistics, Mayer's method of averages, multiple linear regression, Régnier's transfer algorithm |
| Topics |
Classification - Clustering - Partitioning, History of Statistics, Statistics |
| Abstract |
The simple case of Mayer's straight line fitting, which was taught in French secondary schools some years ago, was introduced in some curricula as a surrogate to least-squares. It turns out that the procedure thus proposed to secondary school students provides a basic example of a regression tree. It also turns out, in the general case, that it is a clustering problem for which Régnier's transfer algorithm [1965] is well suited, albeit possibly suboptimal. The famous example of fitting which Mayer treated in 1750 by an innovative and general method is revisited in the light of standard present-day statistical methods. The numerical results show the outstanding expertise of Mayer. |
| Number |
187, Fall 2009, special issue: 2007 Meeting of the French-speaking Society of Classification |
| Language |
French | Read the article
| Title |
Comparing various approaches of supervised evaluation |
| Author |
FERRANDIZ Sylvain |
| Keywords |
Bayesianism, Description length, Nearest neighbor, Structural risk, Supervised classification |
| Topics |
Classification - Clustering - Partitioning, Statistics, Test |
| Abstract |
Instance selection for the nearest neighbor rule is a classical topic in statistical learning. Within the context of hypothesis selection, the characteristics of this problem is that: the set of hypotheses is structured and depends on the data. We thus propose specific nonparametric criteria. We aim at comparing sets of instances of varying size without introducing an extra parameter. Balancing approaches give tools to solve this problem.
Three approaches are considered successively : the SRM (standing for Structural Risk Minimization) approach, the BIC (standing for Bayesian Information Criterion) approach end the MDL (standing for Minimum Description Length) approach. The exploration of each one leads to the definition of a regularized criterion. Each criterion permits the comparison of sets of instances of various size. Each criterion is nonparametric.
We make use of real and synthetic datasets to prove the following point: the MDL criterion is finer than the BIC criterion which, in turn, is finer than the SRM criterion. |
| Number |
187, Fall 2009, special issue: 2007 Meeting of the French-speaking Society of Classification |
| Language |
French | Read the article
| Title |
A new clustering method for interval data |
| Author |
HARDY André, KASORO Nathanael |
| Keywords |
Clustering, Decision tree, Hypervolumes criterion, Maximum likelihood, Poisson Process |
| Topics |
Classification - Clustering - Partitioning, Statistics, Trees |
| Abstract |
This paper presents a new clustering method for interval data. It is an extension of a classical clustering method to interval data. The classical procedure is based on the theory of point processes, and more particularly on the homogeneous Poisson process. The first part of the new method is a monothetic divisive procedure. The cut rule is an extension to interval data of the Hypervolumes clustering criterion. The pruning step uses two statistical likelihood ratio tests based on the homogeneous Poisson process: the Hypervolumes test and the Gap test. The output is a decision tree. The second part of the method is a merging process, that allows in particular cases to improve the classification obtained at the end of the first part of the algorithm. The method is applied to a generated data set and to a real data set. It is compared with other clustering methods available for interval data. |
| Number |
187, Fall 2009, special issue: 2007 Meeting of the French-speaking Society of Classification |
| Language |
French | Read the article
| Title |
Tree construction starting from betweenness relations, application to the stemma codicum |
| Author |
LE POULIQUEN Marc, BARTHELEMY Jean-Pierre |
| Keywords |
Betweenness, Filiation of manuscripts, Stemma codicum, Tree |
| Topics |
Classification - Clustering - Partitioning, Literature, Trees |
| Abstract |
In this paper, we model the ternary betweenness relation within the framework of the critical edition of texts. The editor must try to reconstruct as well as possible, starting from the various preserved manuscripts, the original manuscript as the author wrote it. The corpus is made up of many manuscripts which are copied from one another. To do so, it appears interesting to draw up a family tree of these manuscripts called stemma codicum. A manuscript B is between the manuscripts A and C, i.e. the manuscript C was copied starting from the manuscript B which itself was copied from A. This is this concept of betweenness by copy act which one wishes to model. |
| Number |
187, Fall 2009, special issue: 2007 Meeting of the French-speaking Society of Classification |
| Language |
French | Read the article
| Title |
Text formal reductions for tree analysis and automatic classification. Application to a latin historians corpus |
| Author |
MELLET Sylvie, LUONG Nguyen Xuan, LONGREE Dominique, BARTHELEMY Jean-Pierre |
| Keywords |
Generic classification, Lattice, Linear textual structures, Motif (pattern), Neighbourhood, Texts topological approach, Tree analysis |
| Topics |
Classification - Clustering - Partitioning, Lattices, Literature, Trees |
| Abstract |
In this paper, we present different methods of automatic classification applied to a corpus of literary texts and we compare their different results; in particular we evaluate how each of them is suitable for exhibiting the generic classification of the corpus. We demonstrate that a topological approach of the texts which takes into account their linearity, i.e. the order of their micro- and macro-structures, results in better clustering than traditional quantitative methods which leave generally out of count this linear structure. |
| Number |
187, Fall 2009, special issue: 2007 Meeting of the French-speaking Society of Classification |
| Language |
French | Read the article
Users rights :
The entire journal is licensed under a Creative Commons license
|
