Matches in SemOpenAlex for { <https://semopenalex.org/work/W83271465> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W83271465 endingPage "18" @default.
- W83271465 startingPage "1" @default.
- W83271465 abstract "AbstractClusterwise regression is a technique for clustering data. Instead of using the clas-sical homogeneity or separation criterion, clusterwise regression is based upon theaccuracy of a linear regression model associated to each cluster. This model has manyadvantages, specially for the purpose of data mining, however, the underlying math-ematical model is difficult to solve due to its large number of local optima. In thispaper, we propose the use of the Variable Neighborhood Search metaheuristic (VNS)to improve the quality of the solution. Two perturbation strategies are described andone of them yields a substantial improvement if compared to multistart (the error isreduced by a factor of more than 1.5 on average for the 10 clusters problem).R´esum´eLa r´egression par classes est une technique de classification des donn´ees. Au lieud’utiliser les crit`eres classiques d’homog´en´eit´e et de s´eparation, la r´egression par classesest bas´ee sur l’ad´equation d’un mod`ele de r´egression lin´eaire associ´e a chaque classe.Ce mod`ele a beaucoup d’avantages, particuli`erement pour le data mining. Toutefois,le mod`ele math´ematique sous-jacent est difficile a r´esoudre a cause de son grand nom-bre d’optima locaux. Dans cet article, nous proposons l’utilisation de la recherche avoisinages variables (VNS) pour am´eliorer la qualit´e de la solution. Deux strat´egiesde perturbation sont d´ecrites et une d’elles donne des am´eliorations substantielles parrapport au multistart (l’erreur est r´eduite par un facteur de plus de 1,5 en moyennepour le probl`eme a 10 classes)." @default.
- W83271465 created "2016-06-24" @default.
- W83271465 creator A5038480864 @default.
- W83271465 creator A5040109602 @default.
- W83271465 date "2005-08-01" @default.
- W83271465 modified "2023-09-24" @default.
- W83271465 title "Variable Neighborhood Search for Least Squares Clusterwise Regression" @default.
- W83271465 cites W1521171399 @default.
- W83271465 cites W168241150 @default.
- W83271465 cites W1966322875 @default.
- W83271465 cites W1983989808 @default.
- W83271465 cites W1988423736 @default.
- W83271465 cites W1994267449 @default.
- W83271465 cites W1998140754 @default.
- W83271465 cites W2010334716 @default.
- W83271465 cites W2015392889 @default.
- W83271465 cites W2023689787 @default.
- W83271465 cites W2046121807 @default.
- W83271465 cites W2048085529 @default.
- W83271465 cites W2063960597 @default.
- W83271465 cites W2065537939 @default.
- W83271465 cites W2069120451 @default.
- W83271465 cites W2069657950 @default.
- W83271465 cites W2075594497 @default.
- W83271465 cites W2084812512 @default.
- W83271465 cites W2091292848 @default.
- W83271465 cites W2116675171 @default.
- W83271465 cites W2127218421 @default.
- W83271465 cites W22354473 @default.
- W83271465 cites W2312680915 @default.
- W83271465 cites W2601344776 @default.
- W83271465 hasPublicationYear "2005" @default.
- W83271465 type Work @default.
- W83271465 sameAs 83271465 @default.
- W83271465 citedByCount "0" @default.
- W83271465 crossrefType "journal-article" @default.
- W83271465 hasAuthorship W83271465A5038480864 @default.
- W83271465 hasAuthorship W83271465A5040109602 @default.
- W83271465 hasConcept C105795698 @default.
- W83271465 hasConcept C114614502 @default.
- W83271465 hasConcept C138885662 @default.
- W83271465 hasConcept C152877465 @default.
- W83271465 hasConcept C15708023 @default.
- W83271465 hasConcept C33923547 @default.
- W83271465 hasConcept C48921125 @default.
- W83271465 hasConceptScore W83271465C105795698 @default.
- W83271465 hasConceptScore W83271465C114614502 @default.
- W83271465 hasConceptScore W83271465C138885662 @default.
- W83271465 hasConceptScore W83271465C152877465 @default.
- W83271465 hasConceptScore W83271465C15708023 @default.
- W83271465 hasConceptScore W83271465C33923547 @default.
- W83271465 hasConceptScore W83271465C48921125 @default.
- W83271465 hasLocation W832714651 @default.
- W83271465 hasOpenAccess W83271465 @default.
- W83271465 hasPrimaryLocation W832714651 @default.
- W83271465 hasRelatedWork W168241150 @default.
- W83271465 hasRelatedWork W1971761217 @default.
- W83271465 hasRelatedWork W2004958031 @default.
- W83271465 hasRelatedWork W2023689787 @default.
- W83271465 hasRelatedWork W2029272463 @default.
- W83271465 hasRelatedWork W2063960597 @default.
- W83271465 hasRelatedWork W2065537939 @default.
- W83271465 hasRelatedWork W2090268867 @default.
- W83271465 hasRelatedWork W2126713365 @default.
- W83271465 hasRelatedWork W2174027534 @default.
- W83271465 hasRelatedWork W2312680915 @default.
- W83271465 hasRelatedWork W2962882770 @default.
- W83271465 hasRelatedWork W2963582384 @default.
- W83271465 hasRelatedWork W2963982265 @default.
- W83271465 hasRelatedWork W2989418518 @default.
- W83271465 hasRelatedWork W3021036236 @default.
- W83271465 hasRelatedWork W3112138759 @default.
- W83271465 hasRelatedWork W3123737980 @default.
- W83271465 hasRelatedWork W590635534 @default.
- W83271465 hasRelatedWork W2182226215 @default.
- W83271465 isParatext "false" @default.
- W83271465 isRetracted "false" @default.
- W83271465 magId "83271465" @default.
- W83271465 workType "article" @default.