Computational methods for Traditional Chinese Medicine: A survey

Web Page Classification (WPC) is both an important and challenging topic in data mining. The knowledge of WPC can help users to obtain useable information from the huge internet dataset automatically and efficiently. Many efforts have been made to WPC. However, there is still room for improvement of current approaches. One particular challenge in training classifiers comes from the fact that the available dataset is usually unbalanced. Standard machine learning algorithms tend to be overwhelmed by the major class and ignore the minor one and thus lead to high false negative rate. In this paper, a novel approach for Web page classification was proposed to address this problem by using an ensemble of support vector machine classifiers to perform this work. Principal Component Analysis (PCA) is used for feature reduction and Independent Component Analysis (ICA) for feature selection. The experimental results indicate that the proposed approach outperforms other existing classifiers widely used in WPC.

Web Page Classification using an ensemble of support vector machine classifiers

In this paper, a novel approach of Web page classification using Naive Bayes (NB) classifier based on independent component analysis (ICA) is proposed. In order to perform the classification, a Web page is firstly represented by a vector of features with different weights, and the weight calculated method is improved. As the number of the features is big, principal component analysis (PCA) which is to select the relevant features will perform in preprocessing section as input for improved ICA algorithm (MFICA). Finally, the output of MFICA is sent to NB classifier for classification to boost the classifierpsilas performance. The experimental evaluation demonstrates that the NB classifier based on ICA model provides acceptable classification accuracy.

A Novel Approach to Naive Bayes Web Page Automatic Classification

Computer-aided syndrome differentiation is one of most active issues in how to digitize the Traditional Chinese Medicine. The main purpose of this paper is to show the use of Formal Concept Analysis (FCA) to visualize the dependencies of syndromes of Traditional Chinese Medicine (TCM) with Heart-Spleen syndrome serving as an example. First, the syndromes and syndrome factors are defined as the objects and attributes, then their concepts and the hierarchical relationships are described in mathematical language, finally their hierarchy is visualized by concept lattice. This approach may make syndrome's knowledge clearer, and may make data mining of TCM more efficient.

Using formal concept analysis to visualize relationships of syndromes in traditional chinese medicine

Classification of Web pages is one of the challenging and important task as there is an increase in web pages in day to day life provided by internet. There are many ways of classifying web pages based on different approach and features. This paper explains some of the approaches and algorithms used for the classification of webpages. Web pages are allocated to pre-determined categories which is done mainly according to their content in Web page classification. The important technique for web mining is web page classification because classifying the web pages of interesting class is the initial step of data mining. The agenda of this paper is first to introduce the concepts related to web mining and then to provide a comprehensive review of different classification techniques.

/pdf/comparative-study-of-web-page-classification-approaches-2oln7v467m.pdf

Comparative Study of Web Page Classification Approaches

This paper studies several key aspects of support vector machine (SVM) for Web page classification. Developed from statistical learning theory, SVM is widely investigated and used for text categorization because of its high generalization performance and tolerant ability of processing high dimension classification. Firstly, some methods for Web page presentation are studied. Secondly the Web page classification based on SVM is implementation on data set, and NB classifier is used for study the performance of the SVM classifier processing high dimension space. Finally, the comparison on the polynomial kernel function and the radius basis function (RBF) kernel function is studied. It is proved that if a kernel has a perfect alignment with the classification task, the SVM classifier has better performances

Web Page Classification Based on SVM

In this paper, the Individual Diagnosis and Treatment System (IDTS) based on Chinese Traditional Medicine were studied by using of Data Mining technology in detail. Especially for Coronary disease, by gathering clinical information widely, the IDTS can be established and inconspicuous information could be found, which would be very important and helpful for Chinese Traditional Medicine to improve the clinical diagnosis.

Research and application of data mining in individual diagnosis and treatment based on chinese traditional medicine

Web page classification is an important research direction for web mining. This paper gives some methods for representation of web page, studies on several aspects of support vector machine (SVM) for structural risk minimization concepts of statistical learning theory. The web page classifier and algorithm based on SVM is proposed. Cross validation method is used to select parameters of SVM classifier. The experimental results show that SVM provides an effective method for web page categorization and has promising application in web mining.. . Index Terms—web page categorization, support vector machine, Web Page representation; Kernel function; Web Ming

Weimin Xue

Papers

Web Page Classification Based on SVM

Research and application of data mining in individual diagnosis and treatment based on chinese traditional medicine

Application of SVM in web page categorization