Improving kNN multi-label classification in Prototype Selection scenarios using class proposals
Contribuinte(s) |
Universidad de Alicante. Departamento de Lenguajes y Sistemas Informáticos Reconocimiento de Formas e Inteligencia Artificial |
---|---|
Data(s) |
02/02/2015
02/02/2015
01/05/2015
|
Resumo |
Prototype Selection (PS) algorithms allow a faster Nearest Neighbor classification by keeping only the most profitable prototypes of the training set. In turn, these schemes typically lower the performance accuracy. In this work a new strategy for multi-label classifications tasks is proposed to solve this accuracy drop without the need of using all the training set. For that, given a new instance, the PS algorithm is used as a fast recommender system which retrieves the most likely classes. Then, the actual classification is performed only considering the prototypes from the initial training set belonging to the suggested classes. Results show that this strategy provides a large set of trade-off solutions which fills the gap between PS-based classification efficiency and conventional kNN accuracy. Furthermore, this scheme is not only able to, at best, reach the performance of conventional kNN with barely a third of distances computed, but it does also outperform the latter in noisy scenarios, proving to be a much more robust approach. This work was partially supported by the Spanish Ministerio de Educación, Cultura y Deporte through FPU Fellowship (AP2012–0939), the Spanish Ministerio de Economía y Competitividad through Project TIMuL (TIN2013-48152-C2-1-R), Consejería de Educación de la Comunidad Valenciana through Project PROMETEO/2012/017 and Vicerrectorado de Investigación, Desarrollo e Innovación de la Universidad de Alicante through FPU Program (UAFPU2014–5883). |
Identificador |
Pattern Recognition. 2015, 48(5): 1608-1622. doi:10.1016/j.patcog.2014.11.015 0031-3203 (Print) 1873-5142 (Online) http://hdl.handle.net/10045/44471 10.1016/j.patcog.2014.11.015 |
Idioma(s) |
eng |
Publicador |
Elsevier |
Relação |
http://dx.doi.org/10.1016/j.patcog.2014.11.015 |
Direitos |
info:eu-repo/semantics/openAccess |
Palavras-Chave | #K-Nearest Neighbor #Multi-label classification #Prototype Selection #Class proposals #Lenguajes y Sistemas Informáticos |
Tipo |
info:eu-repo/semantics/article |