SED: supervised experimental design and its application to text classification

In recent years, active learning methods based on experimental design achieve state-of-the-art performance in text classification applications. Although these methods can exploit the distribution of unlabeled data and support batch selection, they cannot make use of labeled data which often carry useful information for active learning. In this paper, we propose a novel active learning method for text classification, called supervised experimental design (SED), which seamlessly incorporates label information into experimental design. Experimental results show that SED outperforms its counterparts which either discard the label information even when it is available or fail to exploit the distribution of unlabeled data. Categories and Subject Descriptors G.3 [Mathematics of Computing]: Probability and Statistics--Experimental Design; H.3 [Information Storage and Retrieval]: Information Search and Retrieval--Clustering General Terms Algorithms, Theory Keywords Active Learning, Supervised ...

• Type: Conference
• Year: 2010
• Where: SIGIR
• Authors: Yi Zhen, Dit-Yan Yeung
• Keywords: Active Learning, Information Technology, SIGIR 2010, Text Classification, Unlabeled Data