Title :
Co-EM Support Vector Machine Based Text Classification from Positive and Unlabeled Examples
Author :
Bang-zuo Zhang ; Wan-li Zuo
Author_Institution :
Coll. of Comput. Sci. & Technol., Jilin Univ., Changchun
Abstract :
This paper has brought about a novel method based on multi-view algorithms for learning from positive and unlabeled examples (LPU). First we, with an improved 1-DNF method, split the text feature into a positive feature set (PF) and a negative feature set (NF). And we project each text vector on the two feature sets in turn. Then we use the co-EM SVM algorithm, which was previously used for semi-supervised learning. Finally, we select the better classifier for the result. Comprehensive evaluation has been performed on the Reuers-21578 collection which shows that our method is efficient and effective.
Keywords :
learning (artificial intelligence); pattern classification; support vector machines; text analysis; multiview algorithms; negative feature set; positive examples; positive feature set; semisupervised learning; support vector machine; text classification; unlabeled examples; Bayesian methods; Educational institutions; Intelligent networks; Machine learning; Noise measurement; Semisupervised learning; Support vector machine classification; Support vector machines; Text categorization; Yield estimation; Co-EM Support Vector Machine; Learning from Positive and Unlabeled examples; Machine Learning; Semi-Supervised Learning; Text classification;
Conference_Titel :
Intelligent Networks and Intelligent Systems, 2008. ICINIS '08. First International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-0-7695-3391-9
Electronic_ISBN :
978-0-7695-3391-9
DOI :
10.1109/ICINIS.2008.29
