Classifying Data Sets Using Support Vector Machines Based on Geometric Distance

Support vector machines （SVMs） are not as favored for large-scale data mining as for pattern recognition and machine learning because the training complexity of SVMs is highly dependent on the size of data set. This paper presents a geometric distance-based SVM （GDB-SVM）. It takes the distance between a point and classified hyperplane as classification rule,and is designed on the basis of theoretical analysis and geometric intuition. Experimental code is derived from LibSVM with Microsoft Visual C ＋＋ 6.0 as system of translating and editing. Four predicted results of five of GDB-SVM are better than those of the method of one against all （OAA）. Three predicted results of five of GDB-SVM are better than those of the method of one against one （OAO）. Experiments on real data sets show that GDB-SVM is not only superior to the methods of OAA and OAO, but highly scalable for large data sets while generating high classification accuracy.

A general framework for progressive point-sampled geometry

Recently unstructured dense point sets have become a new representation of geometric shapes. In this paper we introduce a novel framework within which several usable error metrics are analyzed and the most basic properties of the pro- gressive point-sampled geometry are characterized. Another distinct feature of the proposed framework is its compatibility with most previously proposed surface inference engines. Given the proposed framework, the performances of four representative well-reputed engines are studied and compared.

Integrated Semantic Similarity Model Based on Ontology

To solve the problem of the inadequacy of semantic processing in the intelligent question answering system, an integrated semantic similarity model which calculates the semantic similarity using the geometric distance and information content is presented in this paper. With the help of interrelationship between concepts, the information content of concepts and the strength of the edges in the ontology network, we can calculate the semantic similarity between two concepts and provide information for the further calculation of the semantic similarity between user’s question and answers in knowledge base. The results of the experiments on the prototype have shown that the semantic problem in natural language processing can also be solved with the help of the knowledge and the abundant semantic information in ontology. More than 90% accuracy with less than 50 ms average searching time in the intelligent question answering prototype system based on ontology has been reached. The result is very satisfied. Key words intelligent question answering system - ontology - semantic similarity - geometric distance - information content CLC number TP39 Foundation item: Supported by the important science and technology item of China of “The 10th Five-year Plan” (2001BA101A05-04)Biography: LIU Ya-jun (1953-), female, Associate professor, research direction: software engineering, information processing, data-base application.

A Research Review of Conic Fitting for the Plane Data Points

Curve modeling is one of the basic work in computer aided geometric design and computer graphics. For the implicit conic fitting problem in this paper, the research methods that the objective function based on the minimal algebraic distance and geometric distance are summarized. The advantages and disadvantages of every method are analyzed simply, and the applications of the conic fitting are listed.

Distance geometric constraint filtering algorithm and its application in UWB location

This paper proposed a novel wireless location algorithm based on distance geometry （DG） constraint filtering for the time of arrival （TOA） of the signal （namely as DG-TOA）. Filtering and processing of the observed data and leading to the mathematical formulas based on DG-TOA algorithm are applied to location, also play crucial rules. Simulation results show that the proposed DG-TOA algorithm can provide more valid observation data and be more precise than least square estimate （LSE） algorithm in dense, multi-route, indoor circumstances with the ranging estimation error.