Abstract | ||
---|---|---|
Filtering (or editing) is mainly effective in improving the classification accuracy of the Nearest Neighbour (NN) rule, and also in reducing its storage and computational requirements. This work reviews some well-known editing algorithms for NN classification and presents alternative approaches based on combining the NN and the Nearest Centroid Neighbourhood of a sample. Finally, an empirical analysis over real data sets is provided. |
Year | DOI | Venue |
---|---|---|
2002 | 10.1007/3-540-36079-4_21 | CCIA |
Keywords | Field | DocType |
nearest neighbour,training prototypes,computational requirement,nearest centroid neighbourhood,empirical analysis,classification accuracy,nearest neighbour classification,nn classification,well-known editing algorithm,alternative approach | Data mining,Nearest neighbour,Data set,Pattern recognition,Computer science,Computer data storage,Filter (signal processing),Neighbourhood (mathematics),Knowledge extraction,Artificial intelligence,Centroid | Conference |
Volume | ISSN | ISBN |
2504 | 0302-9743 | 3-540-00011-9 |
Citations | PageRank | References |
4 | 0.51 | 29 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
José Salvador Sánchez | 1 | 565 | 31.62 |
R Barandela | 2 | 558 | 23.46 |
F J. Ferri | 3 | 293 | 22.43 |