Elements of information theory
Elements of information theory
COLT '92 Proceedings of the fifth annual workshop on Computational learning theory
A sequential algorithm for training text classifiers
SIGIR '94 Proceedings of the 17th annual international ACM SIGIR conference on Research and development in information retrieval
Selective Sampling Using the Query by Committee Algorithm
Machine Learning
Constrained K-means Clustering with Background Knowledge
ICML '01 Proceedings of the Eighteenth International Conference on Machine Learning
Less is More: Active Learning with Support Vector Machines
ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
Query Learning with Large Margin Classifiers
ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
Support Vector Machine Active Learning with Application sto Text Classification
ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
Clustering with Instance-level Constraints
ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
In Defense of One-Vs-All Classification
The Journal of Machine Learning Research
Class imbalances versus small disjuncts
ACM SIGKDD Explorations Newsletter - Special issue on learning from imbalanced datasets
Learning and evaluating classifiers under sample selection bias
ICML '04 Proceedings of the twenty-first international conference on Machine learning
Learning a Mahalanobis Metric from Equivalence Constraints
The Journal of Machine Learning Research
Batch mode active learning and its application to medical image classification
ICML '06 Proceedings of the 23rd international conference on Machine learning
YALE: rapid prototyping for complex data mining tasks
Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining
Learning on the border: active learning in imbalanced data classification
Proceedings of the sixteenth ACM conference on Conference on information and knowledge management
Hierarchical sampling for active learning
Proceedings of the 25th international conference on Machine learning
Reducing class imbalance during active learning for named entity annotation
Proceedings of the fifth international conference on Knowledge capture
SMOTE: synthetic minority over-sampling technique
Journal of Artificial Intelligence Research
Exploratory undersampling for class-imbalance learning
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Greedy is not Enough: An Efficient Batch Mode Active Learning Algorithm
ICDMW '09 Proceedings of the 2009 IEEE International Conference on Data Mining Workshops
| Hi-index | 0.00 |
We consider the problem of generating balanced training samples from an unlabeled data set with an unknown class distribution. While random sampling works well when the data is balanced, it is very ineffective for unbalanced data. Other approaches, such as active learning and cost-sensitive learning, are also suboptimal as they are classifier-dependent, and require misclassification costs and labeled samples. We propose a new strategy for generating training samples which is independent of the underlying class distribution of the data and the classifier that will be trained using the labeled data. Our methods are iterative and can be seen as variants of active learning, where we use semi-supervised clustering at each iteration to perform biased sampling from the clusters. Several strategies are provided to estimate the underlying class distributions in the clusters and increase the balancedness in the training samples. Experiments with both highly skewed and balanced data from the UCI repository and a private data show that our algorithm produces much more balanced samples than random sampling or uncertainty sampling. Further, our sampling strategy is substantially more efficient than active learning methods. The experiments also validate that, with more balanced training data, classifiers trained with our samples outperform classifiers trained with random sampling or active learning.