Kernel Based Detection of Mislabeled Training Examples

The problem of identifying mislabeled training examples has been examined in several studies, with a variety of approaches developed for editing the training data to obtain better classifiers. Many of these approaches involve applying an individual or an ensemble of classifiers to the training set and filtering the mislabeled examples based on their consistency with respect to the classifier’s outputs. In this study, we formulate mislabeled detection as an optimization problem and introduce a kernel-based approach for filtering the mislabeled examples. Experimental results using a variety of data sets from the UCI data repository demonstrate the effectiveness of our proposed method, compared to existing nearest-neighbor and ensemble-based filtering schemes.

• Type: Conference
• Year: 2007
• Where: SDM
• Authors: Hamed Valizadegan, Pang-Ning Tan
• Keywords: Data Mining, SDM 2007, Training Data, Training Examples, UCI Data Repository