Nearest Neighbour Classifiers for Streaming Data with Delayed Labelling

Author

Kuncheva, Ludmila I. ; Sanchez, J.S.

Author_Institution

Sch. of Electron. & Comput. Sci., Univ. of Wales, Bangor

fYear

2008

fDate

15-19 Dec. 2008

Firstpage

869

Lastpage

874

Abstract

We study streaming data where the true labels come with a delay. The question is whether the online nearest neighbour classifier (IB2 and IB3 here) should employ the unlabelled data. Three strategies are examined: do-nothing, replace and forget. Experiments with 28 data sets show that IB2 benefits from unlabelled data, while IB3 does not.

Keywords

data handling; learning (artificial intelligence); pattern classification; IB2 algorithm; IB3 algorithm; data streaming; delayed labelling; instance-based learning; online nearest neighbour classifier; Computer science; Data mining; Decision support systems; Delay; Demography; Error correction; Labeling; Remuneration; Semisupervised learning; Shape; Pattern recognition; delayed labelling; incremental learning; nearest neighbour classifier; online learning; streaming data;

fLanguage

English

Publisher

ieee

Conference_Titel

Data Mining, 2008. ICDM '08. Eighth IEEE International Conference on

Conference_Location

Pisa

ISSN

1550-4786

Print_ISBN

978-0-7695-3502-9

Type

conf

DOI

10.1109/ICDM.2008.33

Filename

4781193

Link To Document

https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=2710628