Learning from partially supervised data using mixture models and belief functions

Authors:
E. Côme;L. Oukhellou;T. Denux;P. Aknin
Affiliations:
Institut National de Recherche sur les Transports et leur Sécurité (INRETS) - LTN, 2 av. Malleret-Joinville, 94114 Arcueil Cedex, France and Centre de Recherches de Royallieu, Universit& ...;Institut National de Recherche sur les Transports et leur Sécurité (INRETS) - LTN, 2 av. Malleret-Joinville, 94114 Arcueil Cedex, France and Université Paris XII - CERTES, 61 av. du ...;Centre de Recherches de Royallieu, Université de Technologie de Compiègne - HEUDIASYC, B.P. 20529, 60205 Compiègne Cedex, France;Institut National de Recherche sur les Transports et leur Sécurité (INRETS) - LTN, 2 av. Malleret-Joinville, 94114 Arcueil Cedex, France
Venue:
Pattern Recognition
Year:
2009

Citing 21
Cited 11

Uniqueness of information measure in the theory of evidence

Fuzzy Sets and Systems - Special Issue: Measures of Uncertainty
The transferable belief model

Artificial Intelligence
Text Classification from Labeled and Unlabeled Documents using EM

Machine Learning - Special issue on information retrieval
Uncertainty-Based Information: Elements of Generalized Information Theory

Uncertainty-Based Information: Elements of Generalized Information Theory
Estimating a Kernel Fisher Discriminant in the Presence of Label Noise

ICML '01 Proceedings of the Eighteenth International Conference on Machine Learning
Transductive Inference for Text Classification using Support Vector Machines

ICML '99 Proceedings of the Sixteenth International Conference on Machine Learning
Decision making on the sole basis of statistical likelihood

Artificial Intelligence
Semi-supervised learning with an imperfect supervisor

Knowledge and Information Systems
A Boosting Approach to remove Class Label Noise

HIS '05 Proceedings of the Fifth International Conference on Hybrid Intelligent Systems
Classification in the presence of class noise using a probabilistic Kernel Fisher method

Pattern Recognition
Pruning belief decision tree methods in averaging and conjunctive approaches

International Journal of Approximate Reasoning
Bayesian Regularization for Normal Mixture Estimation and Model-Based Clustering

Journal of Classification
Least committed basic belief density induced by a multivariate Gaussian: Formulation with applications

International Journal of Approximate Reasoning
Decision trees as possibilistic classifiers

International Journal of Approximate Reasoning
Belief functions on real numbers

International Journal of Approximate Reasoning
On the plausibility transformation method for translating belief function models to probability models

International Journal of Approximate Reasoning
On information regularization

UAI'03 Proceedings of the Nineteenth conference on Uncertainty in Artificial Intelligence
Partially supervised learning by a credal EM approach

ECSQARU'05 Proceedings of the 8th European conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty
Learning from ambiguously labeled examples

IDA'05 Proceedings of the 6th international conference on Advances in Intelligent Data Analysis
Classification Using Belief Functions: Relationship Between Case-Based and Model-Based Approaches

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Target identification based on the transferable belief model interpretation of dempster-shafer model

IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans

Extending stochastic ordering to belief functions on the real line

Information Sciences: an International Journal
Learning from data with uncertain labels by boosting credal classifiers

Proceedings of the 1st ACM SIGKDD Workshop on Knowledge Discovery from Uncertain Data
Noiseless Independent Factor Analysis with Mixing Constraints in a Semi-supervised Framework. Application to Railway Device Fault Diagnosis

ICANN '09 Proceedings of the 19th International Conference on Artificial Neural Networks: Part II
Knitted fabric defect classification for uncertain labels based on Dempster-Shafer theory of evidence

Expert Systems with Applications: An International Journal
Maximum likelihood estimation from fuzzy data using the EM algorithm

Fuzzy Sets and Systems
Nonparametric criteria for supervised classification of fuzzy data

International Journal of Approximate Reasoning
An evidential reasoning based classification algorithm and its application for face recognition with class noise

Pattern Recognition
Random subspace evidence classifier

Neurocomputing
Variational learning of a Dirichlet process of generalized Dirichlet distributions for simultaneous clustering and feature selection

Pattern Recognition
Estimating mutual information for feature selection in the presence of label noise

Computational Statistics & Data Analysis
Naive possibilistic classifiers for imprecise or uncertain numerical data

Fuzzy Sets and Systems

Quantified Score

Hi-index	0.01

Visualization

Abstract

This paper addresses classification problems in which the class membership of training data are only partially known. Each learning sample is assumed to consist of a feature vector x"i@?X and an imprecise and/or uncertain ''soft'' label m"i defined as a Dempster-Shafer basic belief assignment over the set of classes. This framework thus generalizes many kinds of learning problems including supervised, unsupervised and semi-supervised learning. Here, it is assumed that the feature vectors are generated from a mixture model. Using the generalized Bayesian theorem, an extension of Bayes' theorem in the belief function framework, we derive a criterion generalizing the likelihood function. A variant of the expectation maximization (EM) algorithm, dedicated to the optimization of this criterion is proposed, allowing us to compute estimates of model parameters. Experimental results demonstrate the ability of this approach to exploit partial information about class labels.