Publications du laboratoire
(22) Production(s) de JACQUES J.


ordinalClust: a package for analyzing ordinal data
Auteur(s): Selosse M., Jacques J., Biernacki Christophe
(Document sans référence bibliographique)
Ref HAL: hal01678800_v1
Résumé: Ordinal data are used in a lot of domains, especially when measurements are collected from persons by observations, testings, or questionnaires. ordinalClust is an R package dedicated to ordinal data that proposes tools for modeling, clustering, coclustering and classification. Ordinal data are modeled by the BOS distribution, which is a meaningful model parametrized by a position and a precision parameter. On one hand, the coclustering framework uses the Latent Block Model (LBM) and an SEMGibbs algorithm for the parameters inference. On the other hand, the clustering and the classification methods follow on from simplified versions of this algorithm. An overview of these methods is given, and the way of using them with the ordinalClust package is described through real datasets.



Clustering multivariate functional data in groupspecific functional subspaces
Auteur(s): Schmutz A., Jacques J., Bouveyron Charles, Cheze Laurence, Martin Pauline
(Document sans référence bibliographique)
Ref HAL: hal01652467_v1
Résumé: With the emergence of numerical sensors in many aspects of everyday life, there is an increasing need in analyzing multivariate functional data. This work focuses on the clustering of those functional data, in order to ease their modeling and understanding. To this end, a novel clustering technique for multivariate functional data is presented. This method is based on a functional latent mixture model which fits the data in groupspecific functional subspaces through a multivariate functional principal component analysis. A family of parsimonious models is obtained by constraining model parameters within and between groups. An EMlike algorithm is proposed for model inference and the choice of hyperparameters is addressed through model selection. Numerical experiments on simulated datasets highlight the good performance of the proposed methodology compared to existing work. This algorithm is then applied for analyzing the pollution in U.S. cities for one year.



Analyzing health quality survey using constrained coclustering model for ordinal data and some dynamic implication
Auteur(s): Selosse M., Jacques J., Biernacki Christophe, CoussonGélie Florence
(Document sans référence bibliographique)
Ref HAL: hal01643910_v1
Résumé: The dataset which motivated this work is a psychological survey on women affected by a breast tumor. Patients replied at different moments of their treatment to questionnaires with answers on ordinal scale. The questions relate to aspects of their life called dimensions. To assist the psychologists in analyzing the results, it is useful to emphasize a structure in the dataset. The clustering method achieves that by creating groups of individuals that are depicted by a representative of the group. From a psychological position , it is also useful to observe how questions may be grouped. This is why a clustering should be performed also on the features, which is called a coclustering problem. However , gathering questions that are not related to the same dimension does not make sense from a psychologist stance. Therefore, a constrained coclustering has been performed to prevent questions from different dimensions from getting assembled in a same columncluster. Then, evolution of coclusters along time has been investigated. The method relies on a constrained Latent Block Model embedding a probability distribution for ordinal data. Parameter estimation relies on a Stochastic EMalgorithm associated to a Gibbs sampler, and the ICLBIC criterion is used for selecting the numbers of coclusters.



Anomaly Prevision in Radio Access Networks Using Functional Data Analysis
Auteur(s): Ben Slimen Y., Allio Sylvain, Jacques J.
Conference: IEEE GlobeCom 2017 (Singapour, SG, 20171204)
Actes de conférence: , vol. p. (2017)
Ref HAL: hal01613475_v1
Résumé: In order to help the network maintainers with the daily diagnosis and optimization tasks, a supervised model for mobile anomalies prevention is proposed. The objective is to detect future malfunctions of a set of cells, by only observing key performance indicators that are considered as functional data. Thus, by alerting the engineers as well as selforganizing networks, mobile operators can be saved from a certain performance degradation. The model has proven its efficiency with an application on real data that aims to detect capacity degradation, accessibility and call drops anomalies for LTE networks.



The Functional Latent Block Model for the CoClustering of Electricity Consumption Curves
Auteur(s): Bouveyron Charles, Bozzi Laurent, Jacques J., Jollois FrançoisXavier
(Document sans référence bibliographique) 20170601
Ref HAL: hal01533438_v1
Résumé: As a consequence of the recent policies for smart meter development, electricity operators are nowadays able to collect data on electricity consumption widely and with a high frequency. This is in particular the case in France where EDF will be able soon to remotely record the consumption of its 27 millions clients every 30 minutes. We propose in this work a new coclustering methodology, based on the functional latent block model (funLBM), which allows to build "summaries" of these large consumption data through coclustering. The funLBM model extends the usual latent block model to the functional case by assuming that the curves of one block live in a lowdimensional functional subspace. Thus, funLBM is able to model and cluster large data set with highfrequency curves. An SEMGibbs algorithm is proposed for model inference. An ICL criterion is also derived to address the problem of choosing the number of row and column groups. Numerical experiments on simulated and original Linky data show the usefulness of the proposed methodology.



Modelbased coclustering for ordinal data
Auteur(s): Jacques J., Biernacki Christophe
Conférence invité: 23th Summer Working Group on ModelBased Clustering of the Department of Statistics of the University of Washington (Paris, FR, 20160718)



Modelbased coclustering for functional data
Auteur(s): Ben Slimen Y., Allio Sylvain, Jacques J.
Conference: 24th International Conference on Computational Statistics (Oviedo, ES, 201608)
Ref HAL: hal01383920_v1
Résumé: A modelbased coclustering algorithm for functional data is presented.This algorithm relies on the latent block model using a Gaussian model for the functionalprincipal components and a SEMGibbs algorithm for inference.
