Applied Mechanics and Materials Vols. 701-702

Abstract: It has great impact on result of the network test or simulation if the test simulated traffic is corresponding to real situation. The network traffic is the superposition of different traffic streams in the actual usage of the network. But because of the complexity and time-consumption to generate different traffic streams, it is difficult to generate the network traffic in the simulation for the large scale network. This paper proposes a kind of method for traffic generating based on genetic algorithm .According to building the self-similar traffic model ,the optimal values of the model’s parameters has been obtained. A case study shows the effectiveness of the method for the network reliability.

Abstract: As a nonparametric classification algorithm, K-Nearest Neighbor (KNN) is very efficient and can be easily realized. However, for large dataset, the computational demands for classifying instances using KNN can be expensive. A way to solve this problem is through the condensing approach. It means we remove instances that will bring computational burden but do not contribute to better classification accuracy. This paper proposes a novel weighted distance KNN algorithm based on instances condensing algorithm. The proposed idea is to extract some representative instances and take the processed result as a new training sample set. Meanwhile, use the distance-weighted WDKNN algorithm to improve the prediction accuracy, our experiments show that the proposed strategy can dramatically shorten the time consumption compared with the traditional KNN. On average, the speedup ratios improve 90% while classification accuracy only has 2% decreases.

Abstract: This paper tries to find a more feasible method to achieve core and reduction. Against concepts "distinguishable relation of attribute set" and "distinguishable unit set of attribute set", it defines a concept "important index", and proposes an effective and quick approach for important index. After drawn out the involved theory and equivalent proposition, also presents algorithms for core and reduction upon the important index. The heuristic reduction algorithm adopts the bottom-up design, and gets reduction based on the heuristic information "important index of attribute set". The complexity of the algorithm in space is O(m), and the complexity in time is O(mn²). The theoretical analysis and results show that the ways proposed here simplify the relevant operations and are suitable to deal with the huge volume of data.

Abstract: It is well known that nonlinear equations systems (NESS) is a subclass of nonlinear optimization problem, it exists in many application fields, such as information industry, network design, mechanics and robotics, etc.. How to design feasible and effective optimization methods to obtain the optimal solution or satisfied precision requirement’s optimal solution for complicated NESS is very important in computation fields. In this paper, each nonlinear sub-equation of NESS is approximately regarded as a sub-objective function of multi-objective optimization problem, then the original nonlinear equations systems is transformed into a multi-objective optimization problem, and the equivalence relation of the solution between the original NESS and the transformed multi-objective optimization problem is given. In order to effectively solve the nonlinear equations systems, a self-adaptive levy mutation operation is proposed, and a multi-objective optimization evolutionary algorithm to solve the nonlinear equations systems was designed. Computer simulations demonstrate the proposed algorithm can not only increase the diversity of evolutionary population but also make the evolution population quickly to approach the optimal solution or satisfied precision requirement’s optimal solution.

Abstract: Two-way merge sort algorithm has a good time efficiency which has been used widely. The sort algorithm can be improved on speed and efficient based on its own potential parallelism via the parallel processing capacity of multi-core processor and the convenient programming interface of OpenMP. The time complexity is improved to O(nlog₂n/TNUM) and inversely proportional to the number of parallel threads. The experiment results show that the improved two-way merge sort algorithm become much more efficient compared to the traditional one.

Abstract: In driver fatigue warning system, it is a very effective method for detecting Driver fatigue state through the driver's facial expressions and body movements. The main content of this article is to detect the two basic states of the eyes opening and closing and presents the LBP texture detection operator. Firstly we get the face image sequences using infrared video and extract the eye region using ADABOOST. The SVM is used in classifying feature vector of the eyes open and closed detecting of driver fatigue. A large number of experimental results show that the proposed method has high detection accuracy and timeliness.

Abstract: Explore Traditional Chinese Medicine prescription based on Local partial least squares (LPLS). Method: mathematical modeling base on LPLS, gain the VIP sorting, loadings Bi plot. Results: seek out the optimized direction of the prescription. Conclusion: the method optimization the compatibility of the dachengqi decoction is feasible and effective.

Abstract: In the study, the method based on the combination of analytic hierarchy process(AHP) and fuzzy theory is proposed to evaluate civil aviation airport security information management. Assessment indexes of civil aviation airport security information management are studied and given, and the hierarchic tree is formulated based on assessment indexes of civil aviation airport security information management. On the basis of describing evaluation indexes for civil aviation airport security information management, and the evaluation model of civil aviation airport security information management is constructed based on fuzzy analytic hierarchy process. Finally, the case is used to testify the effectiveness of the proposed fuzzy analytic hierarchy process method.

Abstract: The genetic algorithm is used to investigate the Chinese postman problem with the constraints of working time span and load capacity for postman on mixed networks. In course of chromosome encoding, the priority-based encoding for serving edges set is used, and the priority and the strategy of “serve while going” in the shortest path is integrated to use in course of decoding. A genetic scheme is designed to solve the capacitated Chinese postman problem without translating arc routing problem into node routing problem. The simulation experiments show that the methods is feasible and efficient, it can easily solve the capacitated Chinese postman problems on all kinds of complex networks.