{"id":1433,"date":"2009-01-30T13:34:38","date_gmt":"2009-01-30T06:34:38","guid":{"rendered":"http:\/\/www.math.itb.ac.id\/?p=1433"},"modified":"2009-01-30T13:34:38","modified_gmt":"2009-01-30T06:34:38","slug":"30-jan-2009-seminar-a-highly-and-efficient-robust-outlier-labeling","status":"publish","type":"post","link":"https:\/\/multisite.itb.ac.id\/math\/2009\/01\/30\/30-jan-2009-seminar-a-highly-and-efficient-robust-outlier-labeling\/","title":{"rendered":"30 Jan 2009: Seminar A HIGHLY AND EFFICIENT\u00a0 ROBUST OUTLIER LABELING"},"content":{"rendered":"

Algebra Research Group FMIPA ITB conduct a seminar entitled by A HIGHLY AND EFFICIENT\u00a0<\/strong>
\nROBUST OUTLIER LABELING<\/strong> on,<\/p>\n

Date<\/strong> :\u00a0Friday, January 30th, 2009
\nTime<\/strong> : 09.00 – 11.00
\nPlace<\/strong> : Study Hall,\u00a0Labtek 3 Matematika, ITB
\nJl. Ganesa 10, Bandung<\/p>\n

This seminar is going to be presented by Prof.\u00a0Maman\u00a0Djauhari.\u00a0<\/strong><\/p>\n

ABSTRACT\u00a0<\/strong><\/p>\n

Minimum covariance determinant (MCD), introduced by\u00a0Rousseeuw\u00a0(1985), is a
\nhighly robust estimation method of multivariate location and scatter. It
\nsatisfies all desired properties such as: it has high breakdown point and
\nasymptotically tends to the maximum value, it is affine-equivariant, and
\nits influence function is bounded. Its fast version algorithm (Rousseeuw
\nand van Driessen, 1999), also called Fast MCD, is the most widely used
\nrobust estimation method in the literature and applications. Nowadays, it
\nbecomes more and more popular after Hubert et al. (2005) introduce the
\nimproved version of Fast MCD which is able to give the solution as close
\nas possible to the global minimum.<\/p>\n

The ultimate goal of Fast MCD is to define a robust\u00a0Mahalanobis\u00a0distance
\nin order to separate all outlier suspects. This step is called outlier
\nlabeling. However, Fast MCD is not apt for large and high dimensional data
\nsets. This is a natural consequence of the use of (1) covariance
\ndeterminant as multivariate dispersion measure and (2)\u00a0Mahalanobis
\ndistance, which involves the inversion of the covariance matrix, to define
\nordering structure in the set of data vectors in the sense of
\ncenter-outward ordering.<\/p>\n

The ability to overcome that obstacle is very challenging. It is
\nindispensable, for example, in bioinformatics, computer intrusion
\ndetection, data mining, genetic engineering, financial industry and many
\nother fields of application. To contribute to that challenging area, in
\nthis work we start by analyzing the structure of Fast MCD. We show that
\nmathematically, all what we need is a measure of multivariate dispersion
\nand a depth function. The latter will lead us to a new frontier of
\nresearch where computational geometry, computer science, linear algebra,
\nand statistics play simultaneously the most important role.<\/p>\n

Later on, we introduce a new theorem on depth function and a new theorem
\non measure of multivariate dispersion to construct a new robust estimation
\nmethod of location and scatter which can be used even when data sets are
\nlarge and of high dimension. Based on these estimates we define a robust
\ncenter-outward ordering and a new outlier labeling. For data sets of low
\ndimension, the new method is computationally far more efficient than Fast
\nMCD and, as simulation experiments strongly indicate, it is as robust as
\nFast MCD. For high dimensional data sets, where Fast MCD does not work, it
\nis able to handle the obstacle mentioned above.<\/p>\n

Keywords: affine-equivariant, breakdown point, covariance determinant,
\ndepth function, influential function, multivariate variability measure\u00a0<\/span><\/p>\n

\u00a0<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

Algebra Research Group FMIPA ITB conduct a seminar entitled by A HIGHLY AND EFFICIENT\u00a0 ROBUST OUTLIER LABELING on, Date :\u00a0Friday, January 30th, 2009 Time : 09.00 – 11.00 Place : Study Hall,\u00a0Labtek 3 Matematika, ITB Jl. Ganesa 10, Bandung This seminar is going to be presented by Prof.\u00a0Maman\u00a0Djauhari.\u00a0 ABSTRACT\u00a0 Minimum covariance determinant (MCD), introduced by\u00a0Rousseeuw\u00a0(1985), […]<\/p>\n","protected":false},"author":1479,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[6,29],"tags":[],"class_list":["post-1433","post","type-post","status-publish","format-standard","hentry","category-algebra","category-events"],"_links":{"self":[{"href":"https:\/\/multisite.itb.ac.id\/math\/wp-json\/wp\/v2\/posts\/1433","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/multisite.itb.ac.id\/math\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/multisite.itb.ac.id\/math\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/multisite.itb.ac.id\/math\/wp-json\/wp\/v2\/users\/1479"}],"replies":[{"embeddable":true,"href":"https:\/\/multisite.itb.ac.id\/math\/wp-json\/wp\/v2\/comments?post=1433"}],"version-history":[{"count":0,"href":"https:\/\/multisite.itb.ac.id\/math\/wp-json\/wp\/v2\/posts\/1433\/revisions"}],"wp:attachment":[{"href":"https:\/\/multisite.itb.ac.id\/math\/wp-json\/wp\/v2\/media?parent=1433"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/multisite.itb.ac.id\/math\/wp-json\/wp\/v2\/categories?post=1433"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/multisite.itb.ac.id\/math\/wp-json\/wp\/v2\/tags?post=1433"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}