Wang Xiaoyue

A Novel Web Pages Classification Model Based on Integrated Ontology

The main existed problem in the traditional text classification methods is can’t use the rich semantic information in training data set. This paper proposed a new text classification model based SUMO (The Suggested Upper Merged Ontology) and WordNet ontology integration. This model utilizes the mapping relations between WordNet synsets and SUMO ontology concepts to map terms in document-words vector space into the corresponding concepts in ontology, forming document-concepts vector space, based this, we carry out a text classification experiment. Experiment results show that the proposed method can greatly decrease the dimensionality of vector space and improve the text classification performance.

A semantic information retrieval system based on KIM

This paper studies the OntoText laboratory research project KIM Platform (Knowledge and Information Management Platform). Through the study, We found that the main KIM technical use of the current ontology and natural language processing areas are more highly regarded of the three open source projects, GATE, Sesame, and Lucene. In addition, this paper translated the KIM system into Chinese, and gave details of the physical search, physical model of search, pre-defined pattern search and keyword search implementation. At last, we analysed the problems of KIM system, and future direction of development.

Using Yahoo! Pipes building Information Warehouse of Intelligence

This paper describes the product of Yahoos cofiring of a RSS tool Yahoo! Pipes. We tried to use this tool to build an Information Warehouse of Intelligence. Article details the information market of the build process, including identification of sources of information, content crawling, filtering, sorting, the results of the output, as well as release and use. We identified the main source of information, intelligence in the field of Blog, and information science-related news RSS sources, domestic and foreign well-known experts and scholars in information science personal website, scientific research information, and information science-related meeting information at home and abroad, domestic and international information science major research institution websites, and information science journal articles related to library materials and other information. The key is that information is not a simple RSS feed, but by Yahoo! Pipes of the Filter module carefully screened to ensure the high quality of information.

Applying RDF Ontologies to Improve Text Classification

Current classification methods are based on the “Bag of Words” (BOW) representation, which only accounts for term frequency in the documents, and ignores important semantic relationships between key terms. In this paper, we proposed a system that uses ontologies and Natural Language Processing techniques to index texts. Traditional BOW matrix is replaced by “Bag of Concepts” (BOC). For this purpose, we developed fully automated methods for mapping keywords to their corresponding ontology concepts. Support Vector Machine a successful machine learning technique is used for classification. Experimental results shows that our proposed method dose improve text classification performance significantly

Combination Methodologies of Text Classifier: Design and Implementation

Support vector machines, one of the most population techniques for classification, have been widely used in many application areas. The kernel parameters setting for SVM in a training process impacts on the classification accuracy. Feature selection is another factor that impacts classification accuracy .The objective of this work is to reduce the dimension of feature vectors, optimizing the parameters to improve the SVM classification accuracy and speed. We present rough set method for feature reduce and a genetic algorithm approach for feature selection and parameters optimization to solve this kind of problem. We tried Reuters 21578 using the proposed method. Experimental results indicate, compared with the traditional methods, our proposed method significantly improves the classification accuracy and has fewer input features for support vector machines.

Building Library better though Library 2.0

This paper discusses a new conception Library 2.0.It starts from the concept of understanding, introduces the technology and the feature of library2.0. And then analysis what the Librarian 2.0 should strive to do in the context of the Library 2.0, and finally according to the different voices against Library 2.0, We should actively face expounded Library 2.0 and should actively promote the application level.