Search

Phrases provide a better foundation for indexing and retrieving documents than individual words. Constituents of phrases make other component words in the phrase less ambiguous than when the words appear separately. Intuitively, classifiers that employ phrases for indexing should perform better than those that use words. Although pioneers have explored the possibility of indexing English documents decades ago, there are relatively fewer similar attempts for Chinese documents, partially because segmenting Chinese text into words correctly is not easy already. We build a domain dependent word list with the help of Chien's PAT tree-based method and HowNet, and use the resulting word list for defining relevant phrases for classifying Chinese judicial documents. Experimental results indicate that using phrases for indexing indeed allows us to classify judicial documents that are closely similar to each other. With a relatively more efficient algorithm, our classifier offers better performances than those reported in related works. [ABSTRACT FROM AUTHOR]

Using Bayesian networks as the representation language for student modeling has become a common practice. Many computer-assisted learning systems rely exclusively on human experts to provide information for constructing the network structures, however. We explore the possibility of applying mutual information-based heuristics and support vector machines to learn how students learn composite concepts, based on students' item responses to test items. The problem is challenging because it is well known that students' performances in taking tests do not reflect their competences faithfully. Experimental results indicate that the difficulty of identifying the true learning patterns varies with the degree of uncertainty in the relationship between students' performances in tests and their abilities in concepts. When the degree of uncertainty is moderate, it is possible to infer the unobservable learning patterns from students' external performances with computational techniques. [ABSTRACT FROM AUTHOR]

This paper presents recently discovered properties of mutual information between concepts and dichotomous test items. The properties generalize some common intuitions for comparing test items, and provide principled foundations for designing item-selection heuristics for student assessments in computer-assisted educational systems. We compare performance profiles achieved by systems that adopt mutual information and the Mahalanobis distance in the assessment task. Experimental results reveal that, all else being equal, the mutual information based methods offer better performance profiles. In addition, experimental results suggest that, when computing mutual information online is considered computationally costly, heuristics that are designed based on our theoretical findings serve as a good delegate for exact mutual information. [ABSTRACT FROM AUTHOR]

Along with the increasing accessibility of the Web, services available on the Web have expanded into several aspects of our lives. As a relatively new comer, Web-based legal services have become another front in the arena. For instance, through such services, people obtain rough pictures of the lawsuits of interest before they decide whether and how to approach their counselors. As a step toward offering online legal services, we propose algorithms for classifying criminal charges in Chinese based on the alleged criminal violations that the district attorney used to prosecute the defendants. The classification is not trivial because some prosecutions involve violations of multiple articles in the criminal law, and we wish to classify lawsuits as precisely as possible. We employ techniques of instance-based classification and introspective learning for the classification task, and achieve satisfactory initial results for classifying charges of larceny and gambling. [ABSTRACT FROM AUTHOR]

This extended abstract summarizes an exploration of how computational techniques may help educational experts identify fine-grained student models. In particular, we look for methods that help us learn how students learn composite concepts. We employ Bayesian networks for the representation of student models, and cast the problem as an instance of learning the hidden substructures of Bayesian networks. The problem is challenging because we do not have direct access to students' competence in concepts, though we can observe students' responses to test items that have only indirect and probabilistic relationships with the competence levels. We apply mutual information and backpropagation neural networks for this learning problem, and experimental results indicate that computational techniques can be helpful in guessing the hidden knowledge structures under some circumstances. [ABSTRACT FROM AUTHOR]