Compartir
Título
Tuning Database-Friendly Random Projection Matrices for Improved Distance Preservation on Specific Data
Autor(es)
Materia
Random projection
Nearest neighbor search
Neighborhood preservation
Dimensionality reduction
Randomized algorithms
Clasificación UNESCO
1203.02 Lenguajes Algorítmicos
1206.10 Matrices
Fecha de publicación
2022
Editor
Springerlink
Citación
López-Sánchez, D., de Bodt, C., Lee, J.A. et al. (2022). Tuning Database-Friendly Random Projection Matrices for Improved Distance Preservation on Specific Data. Appl Intell 52, 4927–4939 . https://doi.org/10.1007/s10489-021-02626-6
Resumen
[EN] Random Projection is one of the most popular and successful dimensionality reduction algorithms for large volumes of data. However, given its stochastic nature, different initializations of the projection matrix can lead to very different levels of performance. This paper presents a guided random search algorithm to mitigate this problem. The proposed method uses a small number of training data samples to iteratively adjust a projection matrix, improving its performance on similarly distributed data. Experimental results show that projection matrices generated with the proposed method result in a better preservation of distances between data samples. Conveniently, this is achieved while preserving the database-friendliness of the projection matrix, as it remains sparse and comprised exclusively of integers after being tuned with our algorithm. Moreover, running the proposed algorithm on a consumer-grade CPU requires only a few seconds.
URI
ISSN
0924-669X
DOI
10.1007/s10489-021-02626-6
Versión del editor
Colecciones
- BISITE. Artículos [290]
Patrocinador
Publicación en abierto financiada por el Consorcio de Bibliotecas Universitarias de Castilla y León (BUCLE), con cargo al Programa Operativo 2014ES16RFOP009 FEDER 2014-2020 DE CASTILLA Y LEÓN, Actuación:20007-CL - Apoyo Consorcio BUCLE
Ficheros en el ítem
Tamaño:
18.05Mb
Formato:
Adobe PDF