FISA: Feature-Based Instance Selection for Imbalanced Text Classification

Lecture notes in computer sciencePublished 1 January 2006

Aixin Sun, Ee‐Peng Lim, Boualem Benatallah, Mahbub Hassan

DOI10.1007/11731139_30

Source URL

Citations6

SJR quartileQ2

SJR score0.35

SNIP0.55

TL;DR

A generic algorithm, known as FISA (Feature-based Instance Selection Algorithm), is proposed to select only a subset of negative training documents for training a SVM classifier, which can be more efficiently trained while delivering comparable or better classification accuracy.

Abstract

Support Vector Machines (SVM) classifiers are widely used in text classification tasks and these tasks often involve imbalanced training. In this paper, we specifically address the cases where negative training documents significantly outnumber the positive ones. A generic algorithm known as FISA (Feature-based Instance Selection Algorithm), is proposed to select only a subset of negative training documents for training a SVM classifier. With a smaller carefully selected training set, a SVM classifier can be more efficiently trained while delivering comparable or better classification accuracy. In our experiments on the 20-Newsgroups dataset, using only 35% negative training examples and 60% learning time, methods based on FISA delivered much better classification accuracy than those methods using all negative training documents.

Keywords

Computer Science

Journal of Artificial Intelligence ResearchSMOTE: Synthetic Minority Over-sampling Technique

31,194 Citations2002Nitesh V. Chawla, Kevin W. Bowyer +2 more

A combination of the method of oversampling the minority (abnormal) class and under-sampling the majority class can achieve better classifier performance (in ROC space) and a combination of these methods and the area under the Receiver Operating Characteristic curve (AUC) and the ROC convex hull strategy is evaluated.

View PDF

TechnometricsIntroduction to Statistical Quality Control

7,900 Citations1992Roger Sauter, Douglas C. Montgomery

An extensive empirical study of feature selection metrics for text classification

2,389 Citations2003George Forman

Addressing the Curse of Imbalanced Training Sets: One-Sided Selection.

2,171 Citations1997Miroslav Kubát, Stan Matwin

One-class svms for document classification

1,225 Citations2002Larry M. Manevitz, Malik Yousef

The SVM approach as represented by Schoelkopf was superior to all the methods except the neural network one, where it was, although occasionally worse, essentially comparable.

IEEE Transactions on Knowledge and Data EngineeringKBA: kernel boundary alignment considering imbalanced data distribution

383 Citations2005Gang Wu, Edward Yi Chang

A kernel-boundary-alignment algorithm is proposed, which considers THE training data imbalance as prior information to augment SVMs to improve class-prediction accuracy and works effectively on several data sets.

ACM SIGKDD Explorations NewsletterExtreme re-balancing for SVMs

285 Citations2004Bhavani Raskutti, Adam Kowalczyk

There is a consistent pattern of performance differences between one and two-class learning for all SVMs investigated, and these patterns persist even with aggressive dimensionality reduction through automated feature selection.

Training text classifiers with SVM on very few positive examples

44 Citations2003Janez Brank, Nataša Milić-Frayling +1 more

This work shows that the linear support vector machine (SVM) learning algorithm is adversely affected by imbalance in the training data, and proposes a method based on the conditional class distributions for SVM scores that works well when very few training examples is available to the learner.

View PDF

Integrating feature and instance selection for text classification

37 Citations2002Dimitris Fragoudis, Dimitris Meretakis +1 more

A new algorithm, which is called FIS (Feature and Instance Selection) that targets both problems simultaneously in the context of text classification and shows that FIS considerably reduces both the number of features and thenumber of instances.

Multi-class SVM with negative data selection for Web page classification

13 Citations2005Chih‐Ming Chen, Hahn-Ming Lee +1 more

A negative data selection method is presented to reduce a large amount of negative data for SVM and keep a high accuracy rate for Web page classification.