Applied Mechanics and Materials Vols. 701-702

Abstract: Query recommendation as an important tool to enhance the user search efficiency has gradually become a hotspot. In the context of big data, using the MapReduce programming model, combined with distributed minimum spanning tree algorithm, a parallel query recommended method based on MapReduce was proposed in this paper. The final results show that the efficiency of query recommendation was greatly improved through parallel computing.

Abstract: Unsupervised Discriminant Projection (UDP) is a typical manifold-based dimensionality reduction method, and has been successfully applied in face recognition. However, UDP suffers from the small sample size problem and usually deteriorates because the basis vectors of UDP are statistically correlated. In order to resolve these problems, we propose an Optimal Uncorrelated Unsupervised Discriminant Projection (OUUDP).The aim of OUUDP is to seek a feature submanifold such that the local scatter is minimized and non-local scatter scatter is maximized simultaneously in the embedding space by using a difference-based optimization objective function. Moreover, we impose an appropriate constraint to make the extracted features statistically uncorrelated. As a result, OUUDP can solve the small sample size problem and exploit statistically uncorrelated features. Experimental results on ORL databases demonstrate the effectiveness of the proposed algorithm.

Abstract: As one of the most popular and effective classification algorithms, Support Vector Machine (SVM) has attracted much attention in recent years. Classifiers ensemble is a research direction in machine learning and statistics, it often gives a higher classification accuracy than the single classifier. This paper proposes a new ensemble algorithm based on SVM. The proposed classification algorithm PB-SVM Ensemble consists of some SVM classifiers produced by PCAenSVM and fifty classifiers trained using Bagging, the results are combined to make the final decision on testing set using majority voting. The performance of PB-SVM Ensemble are evaluated on six datasets which are from UCI repository, Statlog or the famous research. The results of the experiment are compared with LibSVM, PCAenSVM and Bagging. PB-SVM Ensemble outperform other three algorithms in classification accuracy, and at the same time keep a higher confidence of accuracy than Bagging.

Abstract: Study traditional Chinese medicine prescription compatibility chemical components Dose-response relationship based on multiplicative signal correction and partial least squares (MSC-PLS). Method: mathematical modeling base on MSC-PLS. Results: study the compatibility chemical components of the dachengqi decoction; mining the regression coefficient and equation, VIP sorting, loadings Bi plot base on the method. Conclusion: the method mining the data information and optimization the compatibility of the dachengqi decoction cure ileus rats is feasible and effective.

Abstract: Effective detection of software defects is an important activity of software development process. In this paper, we propose an approach to predict residual defects for BOSS project, which applies defect distribution model. Experiment results show that this approach can effectively improve the accuracy of defect prediction.

Abstract: Workspace measuring and positioning system (wMPS) is a kind of large-scale and multi-station intersection system, of which the station layout optimization is a common but important problem. Station optimal topological geometry based on genetic algorithm was proposed in this paper. Firstly, the positioning error, coverage area and cost were taken as objectives to establish the multi-objective optimization function. Secondly, genetic algorithm optimization process was established according to multi-objective function. Thirdly, simulation analysis for layout optimization algorithm of 2-4 stations was performed. The results show that the proposed method is able to quickly converge to optimal solutions, have good adaptability and improve wMPS measuring performance as well.

Abstract: Premature occurring in the process of web log association rule mining by genetic algorithm may lead to no globally optimal solution. In order to avoid such a situation, this paper proposes Web log mining algorithm of association rules based on adaptive genetic strategy, simulated annealing. Based on genetic simulated annealing strategy, the algorithm ensures the diversity of the population from one generation to the application of parallel processing technology. In addition, this algorithm introduces the adaptive crossover probability and mutation probability so as to improve the algorithm's global searching ability. Experimental results show that this algorithm can significantly enhance mining speed and can effectively avoid premature, thus of better strong global convergence.

Abstract: With the rapid development of modern logistics technology and automation of information technology, the AS/RS has been more and more widely used in the field of medicine logistics. In order to more effectively manage AS/RS, improve the rationality of the layout, reduce the number of goods handling and storage costs, it demands to optimize the cargo space allocation for goods .This article adopts multi-objective genetic algorithm to solve the optimization problem of cargo space allocation modeling, and use Jointown pharmaceutical logistics center as an example to further verify the analysis, getting the ideal optimization results and the simulation image.

Abstract: In order to solve the problem of continuous attribute discretization, a new improved SOM clustering algorithm was proposed. The algorithm uses the SOM to achieve the initial cluster and get the clustering up limit, then treats the initial cluster centers as samples and use the BIRCH hierarchical clustering algorithm to get secondary clustering, then solves the problems of inflated clusters and identifies discrete breakpoints set. Finally, find the nearest neighbors of each cluster center among these any samples of Breakpoints sets which belong to its attribute, and use it as a basis of discrete trimming. The experimental results show that the proposed algorithm outperforms the conventional discrete SOM clustering algorithm in the breakpoints set (contour factor to enhance 75%) and discrete accuracy (incompatible degrees closer to 0) aspects.

Abstract: With the rapid development of modern logistics technology and automation information technology, the automation stereoscopic warehouse has been used more and more widely. As an important part of logistics system, it also plays an important role in the field of pharmaceutical logistics. In order to more effectively manage the automation stereoscopic warehouse, and improve the picking efficiency, reduce the cost of goods handling and storage, order picking optimization is required to reduce logistics costs and improve profitability. This paper adopts ant colony algorithm to build model and solve the problems of order picking optimization, and takes the Jointown pharmaceutical logistics center as an example to further verify the analysis and get more ideal optimization path simulation image.