Active learning in cost-sensitive environments

Repository

Active learning in cost-sensitive environments

Show full record

Title: Active learning in cost-sensitive environments
Author: Liu, Alexander Yun-chung
Abstract: Active learning techniques aim to reduce the amount of labeled data required for a supervised learner to achieve a certain level of performance. This can be very useful in domains where unlabeled data is easy to obtain but labelling data is costly. In this dissertation, I introduce methods of creating computationally efficient active learning techniques that handle different misclassification costs, different evaluation metrics, and different label acquisition costs. This is accomplished in part by developing techniques from utility-based data mining typically not studied in conjunction with active learning. I first address supervised learning problems where labeled data may be scarce, especially for one particular class. I revisit claims about resampling, a particularly popular approach to handling imbalanced data, and cost-sensitive learning. The presented research shows that while resampling and cost-sensitive learning can be equivalent in some cases, the two approaches are not identical. This work on resampling and cost-sensitive learning motivates a need for active learners that can handle different misclassification costs. After presenting a cost-sensitive active learning algorithm, I show that this algorithm can be combined with a proposed framework for analyzing evaluation metrics in order to create an active learning approach that can optimize any evaluation metric that can be expressed as a function of terms in a confusion matrix. Finally, I address methods for active learning in terms of different utility costs incurred when labeling different types of points, particularly when label acquisition costs are spatially driven.
Department: Electrical and Computer Engineering
Subject: Active learning Labeled data Supervised learners Utility-based data mining Resampling Cost-sensitive learning Label acquisition
URI: http://hdl.handle.net/2152/7848
Date: 2009-12

Files in this work

Download File: liua72235.pdf
Size: 1.355Mb
Format: application/pdf

This work appears in the following Collection(s)

Show full record


Advanced Search

Browse

My Account

Statistics

Information