Learning Simple Relations: Theory and Applications
P. Berkhin, and J. Becher. In Second SIAM Data Mining Conference, page 420--436. (2002)
Abstract
In addition to classic clustering algorithms, many different approaches to clustering are emerging for objects of special nature. In this article we deal with the grouping of rows and columns of a matrix with non-negative entries. Two rows (or columns) are considered similar if corresponding cross-distributions are close. This grouping is a dual clustering of two sets of elements, row and column indices. The introduced approach is based on the minimization of reduction of mutual information contained in a matrix that represents the relationship between two sets of elements. Our clustering approach contains many parallels with K-Means clustering due to certain common algebraic properties. The obtained resuks have many applications, including grouping of Web visit data.
Description
Learning Simple Relations: Theory and Applications
%0 Conference Paper
%1 Berkhin02learningsimple
%A Berkhin, Pavel
%A Becher, Jonathan
%B In Second SIAM Data Mining Conference
%D 2002
%K 2009 clustering co-clustering relations seminar simple
%P 420--436
%T Learning Simple Relations: Theory and Applications
%U http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.18.6458
%X In addition to classic clustering algorithms, many different approaches to clustering are emerging for objects of special nature. In this article we deal with the grouping of rows and columns of a matrix with non-negative entries. Two rows (or columns) are considered similar if corresponding cross-distributions are close. This grouping is a dual clustering of two sets of elements, row and column indices. The introduced approach is based on the minimization of reduction of mutual information contained in a matrix that represents the relationship between two sets of elements. Our clustering approach contains many parallels with K-Means clustering due to certain common algebraic properties. The obtained resuks have many applications, including grouping of Web visit data.
@inproceedings{Berkhin02learningsimple,
abstract = {In addition to classic clustering algorithms, many different approaches to clustering are emerging for objects of special nature. In this article we deal with the grouping of rows and columns of a matrix with non-negative entries. Two rows (or columns) are considered similar if corresponding cross-distributions are close. This grouping is a dual clustering of two sets of elements, row and column indices. The introduced approach is based on the minimization of reduction of mutual information contained in a matrix that represents the relationship between two sets of elements. Our clustering approach contains many parallels with K-Means clustering due to certain common algebraic properties. The obtained resuks have many applications, including grouping of Web visit data.},
added-at = {2009-12-14T01:12:12.000+0100},
author = {Berkhin, Pavel and Becher, Jonathan},
biburl = {https://www.bibsonomy.org/bibtex/2a9330b3e9056d5b9d9b437ae489ad8c2/r.b.},
booktitle = {In Second SIAM Data Mining Conference},
description = {Learning Simple Relations: Theory and Applications},
interhash = {6e9c5c4609bdc4a1df16bef2b80be808},
intrahash = {a9330b3e9056d5b9d9b437ae489ad8c2},
keywords = {2009 clustering co-clustering relations seminar simple},
pages = {420--436},
timestamp = {2009-12-14T01:12:12.000+0100},
title = {Learning Simple Relations: Theory and Applications},
url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.18.6458},
year = 2002
}