Advanced Materials Research Vols. 989-994

Abstract: Anti-ship cruise missile (ASCM) attack route meets the needs of the attack mission will determine the effect of the subsequent attacks. According to research planning of anti-ship cruise missile route based on improved ANT algorithm by elitist strategy. The first is sensitive factor analysis of route planning based on operational requirements. Second is established an objective function and constraints set of route planning, achieved anti-ship cruise missile route planning model. Finally, based on specific objectives, situation and the user's strategies, it can use improved ANT algorithm to plan anti-ship cruise route.

Abstract: Maritime Operations Command premise is a scientific and efficient assessment of the dynamic and varied sea-battlefield. According to research sea-battlefield situation assessment based on improved intuitionistic fuzzy algorithm based on projection. First is based on intuitionistic fuzzy to establish the sea-battlefield situation information matrix. Second is establishing the sea-battlefield assessment criterion vector. Finally, based on the theory of projection, it can compute the proximity of sea-battlefield situation information matrix and sea-battlefield assessment criterion vector, comprehensive and dynamic assess sea-battlefield in different time slices.

Abstract: As a supervised classification algorithm, Support Vector Machine (SVM) has an excellent ability in solving small samples, nonlinear and high dimensional classification problems. However, SVM is inefficient for imbalanced data sets classification. Therefore, a cost sensitive SVM (CSSVM) should be designed for imbalanced data sets classification. This paper proposes a method which constructed CSSVM based on information entropy, and in this method the information entropies of different classes of data set are used to determine the values of penalty factor of CSSVM.

Abstract: This paper proposes an assembling classifier consisting of a global classifier and a local classifier, named as GCLC. To this end, we present a weighted Support Vector Machine (wSVM) that serves as the global classifier, and a fuzzy k-nearest neighbor (fkNN) that serves as the local one. When a query arrives, wSVM labels it firstly. If the global decision is below some threshold, the local fkNN works to provide an improved decision. Extensive experiments on real datasets demonstrate the performance of GCLC compared with the state of the art.

Abstract: The modal of large flexible structure with low and close modes hardly computes accurate data in the ground. While binocular stereo vision is valued for non-contact measurement, greater flexibility and higher precision in on-orbit identification method. This paper optimized the binocular vision algorithm. The difference between original and new algorithm, which includes the aiming-point number, the identification time and the memory number was shown. The conclusion manifests that new algorithm can improve the identification accuracy and speed up the identification and provide support for on-orbit modal identification.

Abstract: An algorithm of uncertain reasoning which more than one result of a new object can be obtained according to the known knowledge is an important part of an expert system. A new object is an especial decision rule which has only a predecessor. In order to resolve the problem that the differences of attributes’ importance in the new object are not considered in traditional methods of uncertain reasoning, a new uncertain reasoning algorithm based on the rules set which is obtained on the basis of the rough set theory is proposed. In the algorithm, both subjective factors and objective factors in the process of reasoning are considered, and the proportion of subjective factors to objective factors can be controlled by users. So the algorithm is better than the tradition method in flexibility and practicability.

Abstract: The evaluation algorithm is based on the attributes of data objects. There is a certain correlation between attributes, and attributes are divided into key attributes and secondary attributes. This paper proposes an algorithm of attribute reduction based on rough set and the clustering algorithm based on fuzzy set. The algorithm of attributes reduction based on rough set is described in detail first. There are a lot of uncertain data of customer clustering, so traditional method of classification to the incomplete data will be very complex. Clustering algorithm based on fuzzy set can improve the reliability and accuracy of web customers.

Abstract: This article describes a standard document similarity measure, and is used for cultural archives category. And should be used in the classification of cultural archives. the results show that, based on graph model of this representation is valid and feasible.

Abstract: Instruction traces are essential for dynamic analysis in reverse engineering. Code in instruction traces is often obfuscated to hinder analysts from understanding and analyzing in malware and binaries that protected by packers. Non-returning calls and call-stack tampering are two typical kinds of such obfuscation. We propose a deobfuscation approach to fight against these two kinds of obfuscated code. We first apply static analysis on instruction traces to identify obfuscated code. Then we transform obfuscated code into semantically equivalent instructions to make the code be easy to understand. Evaluations results on some packed binaries indicate that our approach works well in deobfuscate instruction traces with non-returning calls and call-stack tampering in high precision.

Abstract: The paper introduces Monte Carlo method and Eugenics genetic algorithm, which be used to generate a great diversity of chaotic attractors firstly. By an analysis of their algorithms, a improved eugenics genetic algorithm is presented to avoid the "genetic drift" phenomenon in attractor graphics. A parameter vector distance limit is adopted to solve the problem and lots of experiments applying equivalent mappings of frieze group are finished to validate effectiveness for algorithm.