Using Weighted Nearest Neighbor to Benefit from Unlabeled Data. - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2006

Using Weighted Nearest Neighbor to Benefit from Unlabeled Data.

Résumé

The development of data-mining applications such as textclassification and molecular profiling has shown the need for machine learning algorithms that can benefit from both labeled and unlabeled data, where often the unlabeled examples greatly outnumber the labeled examples. In this paper we present a two-stage classifier that improves its predictive accuracy by making use of the available unlabeled data. It uses a weighted nearest neighbor classification algorithm using the combined example-sets as a knowledge base. The examples from the unlabeled set are “pre-labeled” by an initial classifier that is build using the limited available training data. By choosing appropriate weights for this pre-labeled data, the nearest neighbor classifier consistently improves on the original classifier.

Dates et versions

hal-01512565 , version 1 (24-04-2017)

Identifiants

Citer

Kurt Driessens, Peter Reutemann, Bernhard Pfahringer, Claire Leschi. Using Weighted Nearest Neighbor to Benefit from Unlabeled Data.. 10th Pacific-Asia Conference on Data Mining (PAKDD 2006), Apr 2006, Singapore, Singapore. pp.60-69, ⟨10.1007/11731139_10⟩. ⟨hal-01512565⟩
76 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More