Paper Info
Title
Semi-supervised multi-class Adaboost by exploiting unlabeled data
Abstract
Semi-supervised learning has attracted much attention in pattern recognition and machine learning. Most semi-supervised learning algorithms are proposed for binary classification, and then extended to multi-class cases by using approaches such as one-against-the-rest. In this work, we propose a semi-supervised learning method by using the multi-class boosting, which can directly classify the multi-class data and achieve high classification accuracy by exploiting the unlabeled data. There are two distinct features in our proposed semi-supervised learning approach: (1) handling multi-class cases directly without reducing them to multiple two-class problems, and (2) the classification accuracy of each base classifier requiring only at least 1/K or better than 1/K (K is the number of classes). Experimental results show that the proposed method is effective based on the testing of 21 UCI benchmark data sets.
Year
DOI
Venue
2011
10.1016/j.eswa.2010.11.062
Expert Syst. Appl.
Keywords
Field
DocType
semi-supervised learning,proposed semi-supervised learning approach,semi-supervised learning algorithm,uci benchmark data set,classification accuracy,naı¨ve bayes classifier,unlabeled data,semi-supervised multi-class adaboost,machine learning,j48 decision tree,multi-class case,semi-supervised learning method,binary classification,multi-class data,decision tree,semi supervised learning,pattern recognition,bayes classifier
Data mining,Semi-supervised learning,One-class classification,Computer science,Unsupervised learning,Artificial intelligence,Learning classifier system,AdaBoost,Pattern recognition,Boosting (machine learning),Linear classifier,Machine learning,Quadratic classifier
Journal
Volume
Issue
ISSN
38
6
Expert Systems With Applications
Citations 
PageRank 
References 
11
0.68
14
Authors
4
Name
Order
Citations
PageRank
Enmin Song117624.53
Dongshan Huang2181.49
Guangzhi Ma3245.32
Chih-Cheng Hung4110.68