• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.11 no.3
• /
• pp.135-142
• /
• 2011

Unlike using the sequence-based representation for a chromosome in previous genetic algorithms for Bayesian structure learning, we proposed a matrix representation-based genetic algorithm. Since a good chromosome representation helps us to develop efficient genetic operators that maintain a functional link between parents and their offspring, we represent a chromosome as a matrix that is a general and intuitive data structure for a directed acyclic graph(DAG), Bayesian network structure. This matrix-based genetic algorithm enables us to develop genetic operators more efficient for structuring Bayesian network: a probability matrix and a transpose-based mutation operator to inherit a structure with the correct edge direction and enhance the diversity of the offspring. To show the outstanding performance of the proposed method, we analyzed the performance between two well-known genetic algorithms and the proposed method using two Bayesian network scoring measures.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.5
• /
• pp.1089-1095
• /
• 2004

The genetic algorithms have been used to take a near optimal solution because The generation of the optimal Steiner tree from a given network is NP-hard problem,. The chromosomes in genetic algorithm are represented with the floating point representation instead of the existing binary string for solving this problem. A spanning tree was obtained from a given network using Prim's algorithm. Then, the new Steiner point was computed using genetic algorithm with the chromosomes in the floating point representation, and it was added to the tree for approaching the result. After repeating these evolving steps, the near optimal Steiner tree was obtained. Using this method, the tree is quickly and exactly approached to the near optimal Steiner tree compared with the existing genetic algorithms using binary string.

• Management Science and Financial Engineering
• /
• v.13 no.1
• /
• pp.35-53
• /
• 2007

We consider a(worst-case) robust optimization version of the Economic Order Quantity(EOQ) model. Order setup costs and inventory carrying costs are assumed to have uncertainty in their values, and the uncertainty description of the two parameters is supposed to be given by an ellipsoidal representation. A genetic algorithm combined with Monte Carlo simulation is proposed to approximate the ellipsoidal representation. The objective function of the model under ellipsoidal uncertainty description is derived, and the resulting problem is solved by another genetic algorithm. Computational test results are presented to show the performance of the proposed method.

• Journal of KIISE:Software and Applications
• /
• v.32 no.4
• /
• pp.268-276
• /
• 2005

This paper deals with optimum communication spanning tree(OCST) problem. Generally, OCST problem is known as NP-hard problem and recently, it is reveled as MAX SNP hard by Papadimitriou and Yannakakis. Nevertheless, many researchers have used polynomial approximation algorithm for solving this problem. This paper uses evolutionary algorithm. Especially, when an evolutionary algorithm is applied to tree network problem such as the OCST problem, representation and genetic operator should be considered simultaneously because they affect greatly the performance of algorithm. So, we introduce a new representation method to improve the weakness of previous representation which is proposed for solving the degree constrained minimum spanning tree problem. And we also propose a new decoding method to generate a reliable tree using the proposed representation. And then, for finding a suitable genetic operator which works well on the proposed representation, we tested three kinds of genetic operators using the information of network or the genetic information of parents. Consequently, we could confirm that the proposed method gives better results than the previous methods.

• Korean Management Science Review
• /
• v.20 no.1
• /
• pp.51-64
• /
• 2003

The JSSP (Job Shop Scheduling Problem) Is one of the most general and difficult of all traditional scheduling problems. The goal of this research is to develop an efficient scheduling method based on genetic algorithm to address JSSP. we design scheduling method based on SGA (Single Genetic Algorithm) and PGA (Parallel Genetic Algorithm). In the scheduling method, the representation, which encodes the job number, is made to be always feasible, initial population is generated through integrating representation and G&T algorithm, the new genetic operators and selection method are designed to better transmit the temporal relationships in the chromosome, and island model PGA are proposed. The scheduling method based on genetic algorithm are tested on five standard benchmark JSSPs. The results were compared with other proposed approaches. Compared to traditional genetic algorithm, the proposed approach yields significant improvement at a solution. The superior results indicate the successful Incorporation of generating method of initial population into the genetic operators.

• Journal of Intelligence and Information Systems
• /
• v.17 no.2
• /
• pp.1-22
• /
• 2011

In this paper, we propose an adaptive genetic algorithm (aGA) approach for effectively solving the sequencing problem with precedence constraints (SPPC). For effective representation of the SPPC in the aGA approach, a new representation procedure, called the topological sort-based representation procedure, is used. The proposed aGA approach has an adaptive scheme using a fuzzy logic controller and adaptively regulates the rate of the crossover operator during the genetic search process. Experimental results using various types of the SPPC show that the proposed aGA approach outperforms conventional competing approaches. Finally the proposed aGA approach can be a good alternative for locating optimal solutions or sequences for various types of the SPPC.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.2
• /
• pp.145-152
• /
• 2004

This paper proposes a representation for the embodiment design of mechanical structures and a genetic algorithm suited for the representation. In order to represent early stages and latter stages of the embodiment design, the designs are modeled as simultaneous multi-objective optimization problems of parametric designs for parts and of layout generation for structures. The study, thus, involves genotypes that are adequate to represent phenotypes of the models for the genetic algorithm to solve the given problems. We demonstrate the implementation of the genetic algorithm with the result applied to the gear equipment design.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.8
• /
• pp.1386-1392
• /
• 2006

This study is a numerical representative modelling analysis for applying the process that unravels networks between cells in genetics to WWW of informatics. Using the probabilistic graphical model, the insight from the data describing biological networks is used for making a probabilistic function. Rather than a complex network of cells, we reconstruct a simple lower-stage model and show a genetic representation level from the genetic based network logic. We made probabilistic graphical models from genetic data and extends them to genetic representation data in the method of network modelling in informatics.

• KSBB Journal
• /
• v.21 no.4
• /
• pp.231-235
• /
• 2006

This study is a numerical representative modeling analysis for the application of the process that unravels networks between cells in genetics to web of informatics. Using the probabilistic graphical model, the insight from the data describing biological networks is used for making a probabilistic function. Rather than a complex network of cells, we reconstruct a simple lower-stage model and show a genetic representation level from the genetic based network logic. We made probabilistic graphical models from genetic data and extends them to genetic representation data in the method of network modeling in informatics

• Journal of KIISE:Software and Applications
• /
• v.32 no.4
• /
• pp.277-289
• /
• 2005

To overcome the long design time/high computational effort/low computational performance of phylogenetic learning featuring selection and reproduction, this paper proposes a genetic representation based on XML. Since genetic programs (GP) and genetic operations of this representation are maintained by the invocation of the built-in off-the-shelf XML parser's API, the proposed approach features significant reduced time consumption of GP design process. Handling only semantically correct GPs with standard XML schema can reduce search space and computational effort. Furthermore, computational performance can be improved by the parallelism of GP caused by the utilization of XML, which is a feasible system and wire format for migration of genetic programs in heterogeneous distributed computer environments. To verify the proposed approach, it is applied to the evolution of social behaviors of multiple agents modeling the predator-prey pursuit problem. The results show that the approach can be applied for fast development and time efficiency of GPs.