• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.49 no.12
• /
• pp.667-674
• /
• 2000

In this paper, a modified real-valued genetic algorithm is developed by using the meiosis for human's chromosome. Unlike common crossover methods adapted in the conventional genetic algorithms, our suggested modified real-valued genetic algorithm makes gametes by conducting the meiosis for individuals composed of chromosomes, and then generates a new individual through crossovers among those. Ultimately, when appling it for the gas data of Box-Jenkin, model and parameter identifications can be concurrently done to construct the optimal model of a neural network in terms of minimizing with the structure and the error.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.1
• /
• pp.116-121
• /
• 2008

This paper introduces a selection operator which utilized similarity and fitness of individuals based on fuzzy inference. Adding similarity feature to fitness, proposed selector obtained the decrease of premature convergence and better performances than other selectors. Moreover, an adoption of steady-state evolution provided enhancement of performances additionally. Experiments of proposed method for deceptive problems were tested and showed better performances than conventional methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.12
• /
• pp.3009-3015
• /
• 2013

Next-generation sequencing (NGS) has enabled whole genome and transcriptome single nucleotide variant (SNV) discovery in cancer and method of the most fundamental being determining an individual's genotype from multiple aligned short read sequences at a position. Bayesian algorithm estimate parameter using posterior genotype probabilities and other method, EM algorithm, estimate parameter using maximum likelihood estimate method in observed data. Here, we propose a novel genotype-calling system and compare and analyze the effect of sample size(S = 50, 100 and 500) on posterior estimate of sequencing error rate, somatic mutation status and genotype probability. The result is that estimate applying Bayesian algorithm even for 50 of small sample size approached real parameter than estimate applying EM algorithm in small sample more accurately.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.11a
• /
• pp.645-648
• /
• 2006

생물학적 데이터 서열에는 크게 DNA 서열과 단백질 서열이 있다. 이들 서열 데이터들은 여러 데이터베이스에 걸쳐 매우 방대한 양을 가지고 있으며, 각각의 서열은 수백 또는 수천 개의 항목들을 가지고 있어 길이가 매우 길다. 일반적으로 유전적인 변형, 또는 변이로부터 보존된 영역이나 특정 패턴들을 서열 안에 포함하고 있는데 생물학적 서열 데이터에서 보존된 영역이나 패턴들은 계통발생학적 근거로 활용 될 수도 있으며 기능과 밀접한 관계를 가지기도 한다. 따라서 서열들로부터 빈번하게 발생하는 패턴을 발견하고자 하는 알고리즘 개발이 요구되고 있다. 초창기 Apriori 알고리즘을 변형하여 빈발 패턴을 발견하고자 하는 노력들로부터 근래에는 PrefixSpan 트리를 이용하여 효과적으로 성능을 개선하고 있지만 아직까지는 여러 번의 데이터베이스 접근이 요구되고 있어 성능저하가 발생한다. 이에 본 논문에서는 접미사 트리를 변형하여 데이터베이스 접근을 획기적으로 줄이고 많은 서열들로부터 빈번하게 발생하는 연속적인 서열을 효과적으로 발견하는 방법을 제안한다.

• Journal of Digital Convergence
• /
• v.19 no.10
• /
• pp.403-410
• /
• 2021

Given an undirected simple graph, k-club is one of the proposed structures to model social groups that exist in various types in Social Network Analysis (SNA). Maximum k-Club Problem (MkCP) is to find a k-club of maximum cardinality in a graph. This paper introduces a Genetic Algorithm called HGA+DROP which can be used to approximate maximum k-club in graphs. Our algorithm modifies the existing k-CLIQUE & DROP algorithm and utilizes Heuristic Genetic Algorithms (HGA) to obtain multiple k-clubs. We experiment on DIMACS graphs for k = 2, 3, 4 and 5 to compare the performance of the proposed algorithm with existing algorithms.

• Journal of KIISE:Software and Applications
• /
• v.30 no.10
• /
• pp.912-924
• /
• 2003

Design restructuring improves software quality by reorganizing design elements and reduces maintenance cost. Object-oriented metrics can help to detect design flaws and find transformations to reorganize design elements. Basically, the transformations must preserve the behavior of an initial system. This paper describes a metric based restructuring technique preserving the behavior of object-oriented designs, founded on set theory, and gives its validity by applying the technique to applications written in Java. This paper also compares the technique with a technique using simulated annealing algorithm to show its effectiveness.

• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.7
• /
• pp.377-383
• /
• 2002

The TSP(Traveling Salesman Problem) has been known as NP-complete, there have been various studies to find the near optimal solution. The nearest neighbor heuristic is more simple than the other algorithms which are to find the optimal solution. This paper designs and implements a new distributed nearest neighbor heuristic with bounding function for the TSP using the master/slave model of PVM(Parallel Virtual Machine). Distributed genetic algorithm obtains a near optimal solution and distributed nearest neighbor heuristic finds an optimal solution for the TSP using the near optimal value obtained by distributed genetic algorithm as the initial bounding value. Especially, we get more speedup using a new genetic operator in the genetic algorithm.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1996.11a
• /
• pp.548-554
• /
• 1996

A new method to locate modular fixtures using an optimization technique is proposed. The optimal fixture arrangement is derived to minimize the elastic deformation of a workpiece. That is, a fixture arrangement is regarded better if it minimized the elastic deformation of the workpiece while fixing a workpart of course. In this approach, the workpiece is projected into two dimensional domain to simplify the 3-dimensional fixture arrangement problem into 2-dimensional one. Thus the problem is reduced to find the optimal positions of one horizontal clamp and three locators which minimize the total deformation of the workpiece and the design variables are the location of the contact points between the boundary of the workpiece and the 4-fixels. The Genetic Algorithm is used for the optimization by mapping each design variables to a gene of a chromosome. The fitness value is the total strain energy of the workpiece calculated by the fin element analysis.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06c
• /
• pp.365-370
• /
• 2008

본 논문에서는 개미 집단 최적화(Ant Colony Optimization, ACO)의 시간적 공간적 효율성을 향상시키기 위해 ACO에 엘리트 콤팩트 유전 알고리즘(Elitist compact Genetic Algorithms, elitist cGAs)의 아이디어를 적용한 콤팩트 개미 최적화(Compact elitist Ant Optimization, CAO)를 제안한다. CAO는 elitist cGAs에서 각 세대마다 염색체의 수를 둘로 고정하고 우월한 염색체를 유지하여 최적의 해를 찾는 방식을 적용하여 개미의 수를 하나로 고정하고 전이 확률식과 페로몬 갱신 규칙을 변형하고 특정 문제에 적용할 수 있는 타부 규칙을 추가한 알고리즘이다. 이 알고리즘의 공간 효율성이 ACO보다 좋다는 것을 증명하고 스테이너 트리 문제(Steiner Tree Problem)에 적용하여 제안된 알고리즘의 시간 효율성이 ACO보다 좋다는 것을 보인다.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.8
• /
• pp.54-61
• /
• 1999

In this paper, we apply one of modified Genetic Algorithms, a Multi-population Genetic Algorithm(MGA) that improves the genetic diversity to determine the fuzzy rule base and the shape of membership functions. The generation of the fuzzy rule base for fuzzy control, generally, depends on expert's experience. We suggest a new evaluation function to optimize fuzzy rule base. Simulation shows that the proposed method has good result.