For the shortcomings in the method of traditional statistics-based feature extraction on PU issues, we put forward feature extraction based on ontology to improve the performance of PU classification. We improved PEBL...
详细信息
For the shortcomings in the method of traditional statistics-based feature extraction on PU issues, we put forward feature extraction based on ontology to improve the performance of PU classification. We improved PEBL algorithm, and get the document vector of positive set using ontology-based feature extraction, then find the strong positive features, which include the crossing semantics in the positive documents and have higher frequency in positive set. The improved algorithm scans the documents twice. First, we get the semantic of the documents by ontology. Second, we filtrate the terms which include none of these semantic to reduce the dimension and obtain the document vector. Experiments had shown that the improved PEBL classifier increases the F1 score by 0.7389%.
In this paper, we propose the basic framework of point- wise topological logic on completely distributive lattices and explore approximate reasoning in it. The logic of this paper is based on pointwise characterizatio...
详细信息
In this paper, we propose the basic framework of point- wise topological logic on completely distributive lattices and explore approximate reasoning in it. The logic of this paper is based on pointwise characterization, therefore the pointwise conception is pervasive. We explore approximate reasoning in abstract logical framework Fl on completely distributive lattice L. We propose the structure of point- wise topological logic F TL , the structure of matching function sigma. and the structure of matching neighborhood group. We investigate approximate reasoning in pointwise topological logic F TL with matching function sigma, develop pointwise topological algorithm of simple approximate reasoning, introduce the essential characteristics of this scheme.
This article gives a capsule view of research on rough set theory and applications ongoing at universities and laboratories in China. Included in this capsule view of rough set research is a brief description of the f...
详细信息
We model the sum and product riddle in public announcement logic, which is interpreted on an epistemic Kripke model. The model is symbolically represented as a finite state program with n agents. A model checking meth...
详细信息
We model the sum and product riddle in public announcement logic, which is interpreted on an epistemic Kripke model. The model is symbolically represented as a finite state program with n agents. A model checking method to the riddle is developed by using the BDD-based symbolic model checking algorithm for logic of knowledge we developed in [7]. The method is implemented by extending the model checker MCTK [7] and then the solution of the riddle is verified successfully.
In this paper, extended Runge‐Kutta‐Nyström type methods for perturbed oscillators with low frequency are presented, which inherit the framework of RKN method and make full use of the special feature of the tru...
In this paper, extended Runge‐Kutta‐Nyström type methods for perturbed oscillators with low frequency are presented, which inherit the framework of RKN method and make full use of the special feature of the true flows. Following the line of ***, *** and ***, we develop a new kind of trees to derive the order conditions for the new methods.
The classical implicit function theorem gives conditions that the function is Fréchet differentiable and the derivative is surjective. In this short article they are generalized to conditions of Lipschitz and mon...
The classical implicit function theorem gives conditions that the function is Fréchet differentiable and the derivative is surjective. In this short article they are generalized to conditions of Lipschitz and monotone type. The newly obtained implicit function theorems are used to derive two sets of sufficient conditions for the existence and uniqueness of solutions to the initial value problems of nonsmooth implicit differential equations.
A biclustering algorithm for gene expressing data is presented. Based on the anti-monotones property of the quality of the data sets with their sizes, the algorithm can get the final biclusters by gradually adding col...
详细信息
A biclustering algorithm for gene expressing data is presented. Based on the anti-monotones property of the quality of the data sets with their sizes, the algorithm can get the final biclusters by gradually adding columns and rows on the data sets. Experimental results show that our algorithm has higher processing speed and quality of clustering than other similar algorithms.
Classification is an essential technology in Pedestrian Detection System (PDS). Until now, single-classifier and basic cascaded classifier had been widely used in PDS;however, most of them can hardly satisfy the 3 req...
详细信息
Classification is an essential technology in Pedestrian Detection System (PDS). Until now, single-classifier and basic cascaded classifier had been widely used in PDS;however, most of them can hardly satisfy the 3 requirements at the same time: high detection speed, high detection rate and low false positive rate. In this paper, we proposed an optimized hierarchical classifier which can satisfy the 3 requirements. The proposed method adopted Corse-to-fine and Early-rejection principles to achieve global high performance. It consists of two hierarchies, the first one is used to quickly reject non-pedestrian objects and select out only a few candidates;the second one makes further verification to these candidates. Furthermore, each hierarchy was optimized with statistical models basing on experiments;and each hierarchy is a treelike classifier which has specific optimization demands. At last, an overall performance evaluation standard is proposed, and the experimental results showed that the proposed classifier had better overall performance.
There exist some issues in current intrusion detection algorithms such as unbalanced detection performance on different types of attacks, and redundant or useless features that will lead to the complexity of detection...
详细信息
There exist some issues in current intrusion detection algorithms such as unbalanced detection performance on different types of attacks, and redundant or useless features that will lead to the complexity of detection model and degradation of detection accuracy. This paper presents an ensemble approach to intrusion detection based on improved multi-objective genetic algorithm. The algorithm generates the optimal feature subsets, which achieve the best trade-off between detection rate and false positive rate through an improved MOGA. And the most accurate and diverse base classifiers are selected to constitute the ensemble intrusion detection model by selective ensemble approach. The experimental results show that the algorithm can solve the feature selection problem of intrusion detection effectively. It can also achieve balanced detection performance on different types of attacks while maintaining high detection accuracy.
Diagrams are used for representing design concepts and constructing function structure in the stage of conceptual design. This paper considers sketches as the language of communication in cooperative conceptual design...
详细信息
Diagrams are used for representing design concepts and constructing function structure in the stage of conceptual design. This paper considers sketches as the language of communication in cooperative conceptual design and presents a tool that enables distributed participants to cooperatively construct diagrams in the form of freehand sketches. The tool is designed and implemented as a multi-agent system, which contains a set of agents to fulfill the requirements, including sketch-based interaction, sketch recognition, communication and coordination. The operating strategies of these agents are also addressed, including communication between agents, sketch recognition algorithm, and semantic consistency maintenance. The proposed approach is implemented in cooperative flowchart drawing.
暂无评论