Based on Jordan curve theorem, a universal classification method based on hyper surface is recently put forward. The experiments show that the new method can efficiently and accurately classify large data size up to 1...
详细信息
Based on Jordan curve theorem, a universal classification method based on hyper surface is recently put forward. The experiments show that the new method can efficiently and accurately classify large data size up to 10/sup 7/ in three-dimensional space. However, the number of training samples needed to design a classifier grows with the dimension of the features. So a way to reduce the dimension of the features without losing any essential information is needed. We put forward a kind of simple and efficient dimension reduction method without losing any essential information to improve the performance of classification based on hyper surface for high dimension data.
The theoiy of the quotient space is a new mathematical tool for the study of the different granularity *** uses a triple(X,f,T)to describe a problem,among which X stands for the domain of the problem,f stands for the ...
详细信息
The theoiy of the quotient space is a new mathematical tool for the study of the different granularity *** uses a triple(X,f,T)to describe a problem,among which X stands for the domain of the problem,f stands for the attribute of the domain,and T stands for the structure of the *** analysis and solution of the problem(X,f,T),along with the further analysis and study of the domain and its structure and attribute,help to the description of the different granularity world based upon the complete *** paper firstly introduces the theory of quotient space,and then focuses on the application of this theoiy through the granularity analysis of the searching in the WWW,which has successfully come to the definite result of different *** about the search engine also are presented.
The quotient space theory obtains a fusion model about semi-order structure under the condition of consistency information, harnessing topology relation among elements of the universe space and hierarchical structure....
详细信息
The quotient space theory obtains a fusion model about semi-order structure under the condition of consistency information, harnessing topology relation among elements of the universe space and hierarchical structure. It has great significance to forming a unified theory structure about information fusion technology. This paper gets a new semi-order structure fusion model under the condition of inconsistency information and proves incompleteness of its semi-order lattice . An example is presented in the paper at last as well as a new method is indicated for Bayesian network structure learning.
Let G be a graph and f: G→ G be a continuous map with at least one periodic point. Using the quote space method, the paper addresses that f is an equicontinuous map if and only if one of the following End(G)+2k+1 con...
详细信息
Let G be a graph and f: G→ G be a continuous map with at least one periodic point. Using the quote space method, the paper addresses that f is an equicontinuous map if and only if one of the following End(G)+2k+1 conditions holds: 1) {f jm(End(G)+2k)!}∞j=1 is uniformly convergent, in which m=1,2,…, End(G)+2k; and 2) There is a positive integer n esuring that {f jn}∞j=1 is uniformly convergent.
Based on the quotient space granular theorem, the image segmentation concept is analyzed and the image segmentation methods are studied, and then the quotient space granular theorem of image segmentation is demonstrat...
详细信息
Based on the quotient space granular theorem, the image segmentation concept is analyzed and the image segmentation methods are studied, and then the quotient space granular theorem of image segmentation is demonstrated. The image segmentation problems are described with triple elements function of the quotient space model (X,f,Γ)⇔([X],[f],[Γ]), according to the quotient space granularity computing, the image segmentation theorem is presented. The methods of images segmentation based on hierarchical and synthesis and combinational technique are exploited and then the segmentation algorithm based granularity synthesis technique is proposed. In this technique, the features of directionality and roughness in texture images data set are firstly extracted respectively to form the different granularities of image, then the different granularity are synthesized according to the theorem of granularity synthesis, finally the texture images is segmented. The experimental results demonstrate that the algorithm is valid for the segmentation of complicated texture images.
Feature extraction or selection is one of the most importmant steps in pattern recognition or pattern classification, data mining, machine learning and so on. In this paper, we introduce the information theory, propos...
详细信息
Principal component analysis (PCA) is an important method in multivariate statistical analysis, and its main idea is compression of dimensionality including variables and samples. In this paper, based on the ideas con...
详细信息
This paper proposes a hierarchical iterative and self-supervised method (HISS) to acquire concept words from a large-scale, un-segmented Chinese corpus. It has two levels of iteration: the EM-CLS algorithm and the Vit...
详细信息
This paper proposes a hierarchical iterative and self-supervised method (HISS) to acquire concept words from a large-scale, un-segmented Chinese corpus. It has two levels of iteration: the EM-CLS algorithm and the Viterbi-C/S algorithm constitute the inner iteration for generating concept words, and the concept word validation constitutes the outer iteration together with the concept word generation. Through multiple iterations, it integrates the concept word generation and validation into a uniform acquisition process. In the process of acquisition, the HISS method can cope with the problem of over-segmentation, over-combination and data sparseness. The experimental result shows that the HISS method is valid for concept word acquisition that can simultaneously increase the precision and recall rate of concept word acquisition.
作者:
Hai ZhugeChina Knowledge Grid Research Group
Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy and Sciences Beijing China
In the human, society, interconnection environment and systems methodology perspectives, this paper answers the following questions: What are the Knowledge Grid and its distinguished features? What are its methodology...
详细信息
In the human, society, interconnection environment and systems methodology perspectives, this paper answers the following questions: What are the Knowledge Grid and its distinguished features? What are its methodology and major research issues? These answers are important to the development of this promising area.
The translation of inheritance nets to default logic has been discussed by Etherington[9],Touretzky[10],*** and inheritance nets are similar in some aspects and based on methods of translating inheritance nets to defa...
详细信息
The translation of inheritance nets to default logic has been discussed by Etherington[9],Touretzky[10],*** and inheritance nets are similar in some aspects and based on methods of translating inheritance nets to default logic,a translation of ontologies to default logic with a priority order on defaults is ***,properties of an ontology and the revision of ontologies can be studied in terms of default *** are assumed to be trees under the subsumption relation between concepts and have deduction rules to infer what are not explicitly *** statements in ontologies are translated to facts of default theories of the ontologies and the default inheritance of properties are represented by normal defaults with a priority order on them due to the intuition that subclasses overriding *** an ontology with a tree structure,it is consistent if and only if the default theory of the ontology has a unique extension.
暂无评论