[ABE-L] Papos Aleatórios - IME-UFF - 08/06 - 14:30

[Patricia]

Boa tarde,
Prezados,

No dia 08 de junho teremos mais uma edição dos Papos Aleatórios com a presença dos docentes Wilson Calmon e Mariana Albi. O resumo e as informações pertinentes a esse papo encontram-se no cartaz em anexo.  Convidamos todos os interessados para essa palestra.
Data: 08/junho/2018 (sexta-feira)Horário: 14:30Local: UFF - Campus do Gragoatá - Bloco G - sala 201
Palestrante: Wilson Calmon e Mariana Albi (ambos do Dep. Estatística - IME-UFF )Título:  ESTIMATING THE NUMBER OF CLUSTERS IN A RANKING DATA CONTEXT
Resumo: Ranking Data abound. Top Ten/Hundred lists abound are plentiful: best [worst] academic journals, movies, cars, footballteams, and so on. Not least, Marden [1] points that transform the continuous data into ranks is an usual strategy adoptedin nonparametric analysis. Ranking Data are essentially multivariate and therefore classical methods of multivariate dataanalysis can be employed to analyse it. In particular, clustering techniques can be applied to identify groups [mainly,not revealed ties] in Ranking Data. Typically, the number of clusters is an input for many clustering procedures and theimportance of the choice of the number of clusters is frequently neglected. However in the last 20 years interesting methodshave been proposed to estimate the number of clusters. For illustration, we could cite the Gap Statistic's Method proposedin Tibshirani et al [2], the Fang and Wang [3] proposal based on clustering stability, the non-parametric approach of Fujita etal [4] or the entropy-based method of Liang et al [5]. However, neither of them was developed for Ranking Data. Our aim isto develop specic methods for estimation of the number of clusters in the Ranking Data context based on the Plackett-LuceModel - [6] and [7]. Particularly, we are investigating a Hierarchical Bayesian approach of Luce Model where the number ofcluster is one hyperparameter.
Referências:[1] Marden, John I. (1995): \Analyzing and Modeling Rank Data" ; Chapman Hall/CRC Monographs on StatisticsApplied Probability, London, 1st Edition.[2] Tibshirani, Robert ; Walther, Guenther; Hastie, Trevor (2001): \Estimating the Number of Clusters in a Data Set Viathe Gap Statistic"; Journal of the Royal Statistical Society Series B (Statistical Methodology) 63(2):411-423.[3] Yixin Fang and Junhui Wang (2012): \Selection of the number of clusters via the bootstrap method"; ComputationalStatistics and Data Analysis, 56, 468-477.[4] Fujita, Andr e; Takahashi, Daniel Y.; Patriota, Alexandre G. (2014): \A non-parametric method to estimate thenumber of clusters"; Computational Statistics and Data Analysis, 73, 27-39.[5] Liang, Jiye; Zhao, Xingwang; Li, Deyu; Cao, Fuyuan; Dang, Chuangyin (2012): \Determining the number of clustersusing information entropy for mixed data"; Pattern Recognition, 45, 2251-2265.[6] Luce, R. D. (1959). Individual choice behavior. New York: Wiley.[7] Plackett, Robert L. (1975): \The Analysis of Permutations."Appl. Statist 24 (2):193-202.
Maiores detalhes dos Papos Aleatórios podem ser vistos em https://sites.google.com/view/paposaleatorios.

Att, Patrícia LusiéProfessora AdjuntaDepartamento de EstatísticaInstituto de Matemática e Estatística - UFF


 
|  | Livre de vírus. www.avg.com.  |

-------------- Próxima Parte ----------
Um anexo em HTML foi limpo...
URL: <https://lists.ime.usp.br/archives/abe/attachments/20180523/24771c60/attachment.html>