Please use this identifier to cite or link to this item:
|Title:||Semi-supervised text classification using partitioned EM|
|Authors:||Cong, G. |
|Keywords:||Labeled and unlabeled data|
|Source:||Cong, G.,Lee, W.S.,Wu, H.,Liu, B. (2004). Semi-supervised text classification using partitioned EM. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2973 : 482-493. ScholarBank@NUS Repository.|
|Abstract:||Text classification using a small labeled set and a large unlabeled data is seen as a promising technique to reduce the labor-intensive and time consuming effort of labeling training data in order to build accurate classifiers since unlabeled data is easy to get from the Web. In  it has been demonstrated that an unlabeled set improves classification accuracy significantly with only a small labeled training set. However, the Bayesian method used in  assumes that text documents are generated from a mixture model and there is a one-to-one correspondence between the mixture components and the classes. This may not be the case in many applications. In many real-life applications, a class may cover documents from many different topics, which violates the one-to-one correspondence assumption. In such cases, the resulting classifiers can be quite poor. In this paper, we propose a clustering based partitioning technique to solve the problem. This method first partitions the training documents in a hierarchical fashion using hard clustering. After running the expectation maximization (EM) algorithm in each partition, it prunes the tree using the labeled data. The remaining tree nodes or partitions are likely to satisfy the one-to-one correspondence condition. Extensive experiments demonstrate that this method is able to achieve a dramatic gain in classification performance. © Springer-Verlag 2004.|
|Source Title:||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Appears in Collections:||Staff Publications|
Show full item record
Files in This Item:
There are no files associated with this item.
checked on Dec 8, 2017
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.