Active Class Selection

Authors:
R. Lomasky;C. E. Brodley;M. Aernecke;D. Walt;M. Friedl
Affiliations:
Computer Science Department, Tufts University, Medford, MA,;Computer Science Department, Tufts University, Medford, MA,;Chemistry Department, Tufts University, Medford, MA,;Chemistry Department, Tufts University, Medford, MA,;Dept. of Geography, Boston University, Boston, MA,
Venue:
ECML '07 Proceedings of the 18th European conference on Machine Learning
Year:
2007

Citing 11
Cited 7

A sequential algorithm for training text classifiers: corrigendum and additional data

ACM SIGIR Forum
Algorithmic stability and sanity-check bounds for leave-one-out cross-validation

COLT '97 Proceedings of the tenth annual conference on Computational learning theory
Classification by pairwise coupling

NIPS '97 Proceedings of the 1997 conference on Advances in neural information processing systems 10
Data mining: practical machine learning tools and techniques with Java implementations

Data mining: practical machine learning tools and techniques with Java implementations
Support vector machine active learning for image retrieval

MULTIMEDIA '01 Proceedings of the ninth ACM international conference on Multimedia
Combining clustering and co-training to enhance text classification using unlabelled data

Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining
Stability and generalization

The Journal of Machine Learning Research
Stopping criterion for boosting based data reduction techniques: from binary to multiclass problem

The Journal of Machine Learning Research
Online Choice of Active Learning Algorithms

The Journal of Machine Learning Research
Class imbalances versus small disjuncts

ACM SIGKDD Explorations Newsletter - Special issue on learning from imbalanced datasets
The class imbalance problem: A systematic study

Intelligent Data Analysis

Why label when you can search?: alternatives to active learning for applying human resources to build classification models under extreme class imbalance

Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining
Active class selection for arousal classification

ACII'11 Proceedings of the 4th international conference on Affective computing and intelligent interaction - Volume Part II
Inductive transfer learning for handling individual differences in affective computing

ACII'11 Proceedings of the 4th international conference on Affective computing and intelligent interaction - Volume Part II
Designing robot learners that ask good questions

HRI '12 Proceedings of the seventh annual ACM/IEEE international conference on Human-Robot Interaction
Studying self- and active-training methods for multi-feature set emotion recognition

PSL'11 Proceedings of the First IAPR TC3 conference on Partially Supervised Learning
Adaptive Selective Learning for automatic identification of sub-kilometer craters

Neurocomputing
New algorithms for budgeted learning

Machine Learning

Quantified Score

Hi-index	0.01

Visualization

Abstract

This paper presents Active Class Selection(ACS), a new class of problems for multi-class supervised learning. If one can control the classes from which training data is generated, utilizing feedback during learning to guide the generation of new training data will yield better performance than learning from any a priorifixed class distribution. ACS is the process of iteratively selecting class proportions for data generation. In this paper we present several methods for ACS. In an empirical evaluation, we show that for a fixed number of training instances, methods based on increasing class stability outperform methods that seek to maximize class accuracy or that use random sampling. Finally we present results of a deployed system for our motivating application: training an artificial nose to discriminate vapors.