Effects of unlabeled data on classification error in normal discriminant analysis

DOI10.1016/j.jspi.2013.11.004zbMath1278.62098OpenAlexW2001271983MaRDI QIDQ389294

Keiji Takai, Kenichi Hayashi

Publication date: 20 January 2014

Published in: Journal of Statistical Planning and Inference (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.jspi.2013.11.004

zbMATH Keywords

asymptotic relative efficiencymissing datanon-normal datasemi-supervised learningnormal discriminationpartially labeled dataunlabeled data

Mathematics Subject Classification ID

Related Items (2)

Asymptotic comparison of semi-supervised and supervised linear discriminant functions for heteroscedastic normal populations ⋮ Finite-sample analysis of impacts of unlabeled data and their labeling mechanisms in linear discriminant analysis

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• High-dimensionality effects in the Markowitz problem and other quadratic programs with linear constraints: risk underestimation
• Robustness of the linear discriminant function to nonnormality: Edgeworth series distribution
• Robustness of the linear discriminant function to nonnormality: Johnson's system
• Text classification from labeled and unlabeled documents using EM
• Improving predictive inference under covariate shift by weighting the log-likelihood function
• Errors of misclassification and their probability distributions when the parent populations are Gompertz
• Errors of misclassification associated with gamma distribution
• Semi-supervised local Fisher discriminant analysis for dimensionality reduction
• Asymptotic Inference with Incomplete Data
• Maximum Likelihood Estimation of the Multivariate Normal Mixture Model
• The General Distribution of the Error Rate of a Classification Procedure with Application to Logistic Regression Discrimination
• The Efficiency of Logistic Regression Compared to Normal Discriminant Analysis
• Normal Discrimination with Unclassified Observations
• The relative value of labeled and unlabeled samples in pattern recognition with an unknown mixing parameter
• Sample selection bias in credit scoring models
• Updating a discriminant function on the basis of unclassified data
• Asymptotic relative efficiency of the linear discriminant function under partial nonrandom classification of the training data
• Robustness of the linear and quadratic discriminant function to certain types of non‐normality
• Estimating the components of a mixture of normal distributions

This page was built for publication: Effects of unlabeled data on classification error in normal discriminant analysis