Beyond Boolean Matrix Decompositions: Toward Factor Analysis and Dimensionality Reduction of Ordinal Data

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989592%3A15310%2F13%3A33147956" target="_blank" >RIV/61989592:15310/13:33147956 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.1109/ICDM.2013.127" target="_blank" >http://dx.doi.org/10.1109/ICDM.2013.127</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICDM.2013.127" target="_blank" >10.1109/ICDM.2013.127</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Beyond Boolean Matrix Decompositions: Toward Factor Analysis and Dimensionality Reduction of Ordinal Data

Popis výsledku v původním jazyce

Boolean matrix factorization (BMF), or decomposition, received a considerable attention in data mining research, both for its direct usefulness in data analysis and its fundamental role in understanding Boolean data. In this paper, we argue that researchshould extend beyond the Boolean case toward more general type of data such as ordinal data. Technically, such extension amounts to replacement of the two-element Boolean algebra utilized in BMF by more general structures, which brings non-trivial challenges. We first present the problem formulation, survey the existing literature, and provide an illustrative example. Second, we present new theorems regarding decompositions of matrices with ordinal data. The theorems helps understand the geometry of decompositions and identify parts of input matrices which are good to focus on when computing factors. Third, we propose two algorithms based on these results along with an experimental evaluation. We conclude the paper with a discussion re

Název v anglickém jazyce

Beyond Boolean Matrix Decompositions: Toward Factor Analysis and Dimensionality Reduction of Ordinal Data

Popis výsledku anglicky

Boolean matrix factorization (BMF), or decomposition, received a considerable attention in data mining research, both for its direct usefulness in data analysis and its fundamental role in understanding Boolean data. In this paper, we argue that researchshould extend beyond the Boolean case toward more general type of data such as ordinal data. Technically, such extension amounts to replacement of the two-element Boolean algebra utilized in BMF by more general structures, which brings non-trivial challenges. We first present the problem formulation, survey the existing literature, and provide an illustrative example. Second, we present new theorems regarding decompositions of matrices with ordinal data. The theorems helps understand the geometry of decompositions and identify parts of input matrices which are good to focus on when computing factors. Third, we propose two algorithms based on these results along with an experimental evaluation. We conclude the paper with a discussion re

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

<a href="/cs/project/GAP103%2F11%2F1456" target="_blank" >GAP103/11/1456: Základy pro zpracování dat s využitím podobností</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2013

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of 2013 IEEE 13th International Conference on Data Mining

ISBN

978-0-7685-5108-2

ISSN

1550-4786

e-ISSN

—

Počet stran výsledku

6

Strana od-do

961-966

Název nakladatele

IEEE Computer Society Press

Místo vydání

Los Alamos

Místo konání akce

Dallas

Datum konání akce

7. 12. 2013

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—