Advanced Materials Research Vol. 717

Abstract: Genetic algorithm is a very important and popular kind of algorithm of evolution computing. In order to use this algorithm better and platform-independently, this paper introduces an implement package which is coded in Java, an object-oriented and platform-independent advanced computer programming language, for genetic algorithm. This package includes several sub-packages. In each sub-package, there are some classes with different roles and functions. After the test, these classes can work properly and efficiently in together. The good effect has been received through using this algorithm in four function optimization problems. For the further goal, some studies even need to be carried out in the future.

Abstract: Shuffled frog leaping algorithm (SFLA) is a meta-heuristic algorithm, which combines the social behavior technique and the global information exchange of memetic algorithms. But the SFLA has the shortcoming of low convergence speed while solving complex optimization problems. Particle swarm optimization (PSO) is a fast searching algorithms, but easily falls into the local optimum for the diversity scarcity of particles. In the paper, a new hybrid optimization called SFLA-PSO is proposed, which introduced PSO to SFLA by combining the fast search strategy of PSO and global search strategy of SFLA. Six benchmark functions are selected to compare the performance of SFLA-PSO, basic PSO, wPSO and SFLA. The simulation results show that the proposed algorithm SFLA-PSO possesses outstanding performance in the convergence speed and the precision of the global optimum solution.

Abstract: Nowadays the functional role of emotions has been recently fully recognized as essential for intelligent systems. In this paper an emotion and behavior model are presented based on the similarity between primary emotion and state machine. A two-layer emotional state generator based on the brain science is introduced firstly. The matrix description of state machine is applied to construct the bottom level of emotion generator. This method could improve the reactive performance of intelligent system. A neural cell model named Lapicque is used to describe the transition of emotion state. Experimental results is presented in the end demonstrate the response advantage of our model.

Abstract: According to the low recognition rate of Hu invariant moments in the target images, this article proposes a vehicle-logo recognition research algorithm based on the modified invariant moments. At first, use the template matching to locate the vehicle-logo rough area and use the edge detection for accurate location. Then, calculate the characteristic value of the modified invariant moments of the vehicle-logo, finally, the vehicle-logo is recognized according to the minimum distance of invariant moments. The experimental results show that the modified invariant moments can improve the recognition rate of target images effectively.

Abstract: Modeling Distance and Direction Relationships (DDR) is a key issue in spatial analysis and spatial reasoning. Various fields such as geology, hydrology, ecology, etc. apply DDR models to help digging out valuable patterns hidden in geoscientific dataset. This paper proposed two quantitative models through a raster-based approach for computing Euclidean distance and cardinal direction relationships, respectively, between a pair of spatial objects in a two-dimensional geographical space. The corresponding algorithms were designed and implemented. This new raster-based modeling can work universally on all types of spatial objects (point, line, polygon, or compound objects) and quantify DDR more accurately due to its sensitivity to object shapes. The usefulness of the modeling was demonstrated by various applications.

Abstract: Ant Colony System (ACS) is a new meta heuristics algorithms to solve hard combinatorial optimization problems. In this paper, we propose hybrid ant colony algotirhm that is searching the second best edge first in the state transition rule and updating the pheromone on edges applying the visited number of edge in the globally best tour. And we evaluate the proposed algorithm according to the maximum time for each trial. The results of a simulation experiment demonstrate that the proposed algorithm is better than, or, at least as good as, that of ACS algorithm in the most sets.

Abstract: Mixed-model assembly lines are widely used in many manufacturing firms to meet diversified demands of consumers without possessing large product inventories. In this paper, we posed order oriented assembly line sequencing as a multiple-objective optimization problem with the objectives to minimize material consumption waviness, the total setup cost, and finished product inventory cost. The multi-objective optimization algorithm based on non-dominated sorting particle swarm optimization (NSPSO) is designed. Computational experiment has been demonstrated to the applicability of using NSPSO to solve the problem and effectiveness of the proposed approach. By means of this research, the valid solutions for order oriented mixed-model assembly line sequence can be offered to the decision makers effectively.

Abstract: Multilevel hypergraph partitioning is an significant and extensively researched problem in combinatorial optimization. Nevertheless, as the primary component of multilevel hypergraph partitioning, coarsening phase has not yet attracted sufficient attention. Meanwhile, the performance of coarsening algorithm is not very satisfying. In this paper, we present a new coarsening algorithm based on multilevel framework to reduce the number of vertices more rapidly. The main contribution is introducing the matching mechanism of weighted inner product and establishing two priority rules of vertices. Finally, the effectiveness of our coarsening algorithm was indicated by experimental results compared with those produced by the combination of different sort algorithms and hMETIS in most of the ISPD98 benchmark suite.

Abstract: The language mixing in multi-language speech recognition is one of the hot issues of concern. After analyzing recognition problem, a method to distinguish language with re-class method according to confidence on multi-language recognition result based on Bayesian decision-making rules with minimum error rate and minimum risk was brought out. It can not only avoid cumbersome language recognition in traditional method but also achieve target of decreasing mixing cognition rate. Experiment on Chinese-English mixing recognition shows that the method can distinguish different language and improve speech recognition rate, which has practicality.

Abstract: In order to make the matrix theory of matrix partial order in multivariate analysis and linear model parameter estimation plays an important application, this paper mainly discusses matrix partial order and its application in linear model to compare. Firstly introduces estimation and model comparison the basis knowledge, then use the matrix partial order theory comparison of generalized ridge estimation and LS estimation. Under the mean square error criterion, discusses the ridge estimation superior to LS estimation problem, On this basis, using Lowner partial order generalized ridge estimators are discussed relative to the LS estimate of good properties, popularize the previous conclusions.