• Journal of the Korea Society of Computer and Information
• /
• v.25 no.11
• /
• pp.41-50
• /
• 2020

Sentimental analysis begins with the search for words that determine the sentimentality inherent in data. Managers can understand market sentimentality by analyzing a number of relevant sentiment words which consumers usually tend to use. In this study, we propose exploring performance of feature selection methods embedded with Particle Swarm Optimization Multi Objectives Evolutionary Algorithms. The performance of the feature selection methods was benchmarked with machine learning classifiers such as Decision Tree, Naive Bayesian Network, Support Vector Machine, Random Forest, Bagging, Random Subspace, and Rotation Forest. Our empirical results of opinion mining revealed that the number of features was significantly reduced and the performance was not hurt. In specific, the Support Vector Machine showed the highest accuracy. Random subspace produced the best AUC results.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2002.11a
• /
• pp.3-7
• /
• 2002

컴퓨터 네트워킹 기술의 발달에 힘입어 분산처리를 이용한 기법이 복잡한 구조물의 최적설계에 널리 사용되고 있다. 최적설계시 구조물이 복잡하고 설계 변수가 많아질수록 설계 변수간의 교호작용이 복잡해지고 국부최적해가 많아지는 특성이 있다. 최근의 최적 설계는 이러한 문제점을 해결하고자 다양한 전역 최적화 기법을 도입하여 적용하고 있다. 본 연구에서는 진화이론을 바탕으로 한 유전자 알고리즘과 실험계획법을 바탕으로 한 반응표면법에 분산처리 기법을 도입하여 인공위성 추진 모듈의 최적화에 적용시켰다. 그 결과 유전자 알고리즘이 조금 더 좋은 최적값을 보였으며 해석시간은 반응표면법을 적용 시켰을 경우가 훨씬 짧았다. 병렬처리 기법을 이용한 위성구조체의 최적설계에 있어 유전자 알고리즘은 해의 전역성에서 반응표면법은 시간의 효율성에서 각각 장점을 보였다.

• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2639-2641
• /
• 2003

유전자 알고리즘은 다윈의 자연선택설과 유전자의 진화 개념을 이용한 적응 탐색 알고리즘으로 적용하고자 하는 문제의 매개 변수를 유전자와 비슷한 데이터 구조로 부호화하고, 유전 연산자를 이용하여 문제의 해답을 찾는 알고리즘이다. 최근 유전자 알고리즘은 이러한 복수개의 목적 함수를 최적화 하기 위한 다중 최적화 문제를 위한 최적화 기술로서의 관심이 크게 다루어지고 있으며 전송 문제, 생산 공정 문제 계획 등과 같은 다목적 함수를 다루는 많은 응용 부분에 대해 적용되고 있다. 본 논문에서는 기본적인 다중 목적 함수용 예와 Gen과 Kim이 제안한 네트워크 신뢰도를 고려한 연결 비용과 메시지 지연을 고려한 이중 구속 통신망 설계 문제를 가지고 가중치 합과 여러 가지 파레토 방법들을 비교하고 연구 검토 하고자 한다.

• The Journal of the Korea Contents Association
• /
• v.15 no.7
• /
• pp.21-29
• /
• 2015

This paper proposes a novel two-tier optimization approach for decision making in many-objective problems. Because the Pareto-optimal solution ratio increases exponentially with an increasing number of objectives, simply finding the Pareto-optimal solutions is not sufficient for decision making in many-objective problems. In other words, it is necessary to discriminate the more preferable solutions from the other solutions. In the proposed approach, user preference-oriented as well as diverse Pareto-optimal solutions can be obtained as candidate solutions by introducing an additional tier of optimization. The second tier of optimization employs the corresponding secondary objectives, global evaluation and crowding distance, which were proposed in previous works, to represent the users preference to a solution and the crowdedness around a solution, respectively. To demonstrate the effectiveness of the proposed approach, decision making for some benchmark functions is conducted, and the outcomes with and without the proposed approach are compared. The experimental results demonstrate that the decisions are successfully made with consideration of the users preference through the proposed approach.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.11
• /
• pp.1106-1110
• /
• 2010

This paper proposes a method of optimizing a stacking strategy for an automated container terminal using multi-objective evolutionary algorithms (MOEAs). Since the yard productivities of seaside and landside are conflicting objectives to be optimized, it is impossible to maximize them simultaneously. Therefore, we derive a Pareto optimal set instead of a single best solution using an MOEA. Preliminary experiments showed that the population is frequently stuck in local optima because of the difficulty of the given problem depending on the yard occupancy rate. To cope with this problem, we propose another method of simultaneously optimizing two problems with different difficulties so that diverse solutions can be preserved in the population. Experimental results showed the proposed method can derive better stacking policies than the compared method solving a single problem given the same computational costs.

• Journal of Navigation and Port Research
• /
• v.35 no.6
• /
• pp.501-507
• /
• 2011

This paper focuses on dispatching strategy for AGVs(Automated Guided Vehicle). The goal of AGV dispatching is assigning AGVs to requested job to minimizing the delay of QCs and the travel distance of AGVs. Due to the high dynamic nature of container terminal environment, the effect of dispatching is hard to predict thus it leads to frequent modification of dispatching decisions. In this situation, approaches based on a single rule are widely used due to its simplicity and small computational cost. However, these approaches have a limitation that cannot guarantee a satisfactory performance for the various performance measures. In this paper, dispatching strategy based on multicriteria heuristics is proposed. The Proposed strategy consists of multiple decision criteria. A multi-objective evolutionary algorithm is applied to optimize weights of those criteria. The result of simulation experiment shows that the proposed approach outperforms single rule-based dispatching approaches.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.3
• /
• pp.375-381
• /
• 2005

Usually genetic algorithms are used to design optimal system. However the performance of the algorithm is determined by the fitness function and the system environment. It is expected that a co-evolutionary algorithm, two populations are constantly interact and co-evolve, is one of the solution to overcome these problems. In this paper we propose three types of co-evolutionary algorithm to solve GA-Hard problem. The first model is a competitive co-evolutionary algorithm that solution and environment are competitively co-evolve. This model can prevent the solution from falling in local optima because the environment are also evolve according to the evolution of the solution. The second algorithm is schema co-evolutionary algorithm that has host population and parasite (schema) population. Schema population supply good schema to host population in this algorithm. The third is game model-based co-evolutionary algorithm that two populations are co-evolve through game. Each algorithm is applied to visual servoing, robot navigation, and multi-objective optimization problem to verify the effectiveness of the proposed algorithms.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.103-106
• /
• 2000

In this paper, we introduce a new evolutionary multiobjective optimization algorithm based on the non-domination direction information, which can be an alternative among several multiobjective evolutionary algorithms. The new evolutionary multiobjective optimization algorithm proposed in this paper will not use the conventional recombination or mutation operators but use the non-domination directions, which are extracted from the non-domination relation among the population. And the problems of the modified sharing algorithms are pointed out and a new sharing algorithm sill be proposed to overcome those problems.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.12a
• /
• pp.229-232
• /
• 2001

Recently, several promising multiobjective evolutionary algorithms, e,g, SPEA, NSGA-ll, PESA, and SPEA2, have been developed. In this paper, we also propose a new multiobjective evolutionary algorithm that compares to them. In the algorithm proposed in this paper, we introduce a novel concept, "inherited age" and total algorithm is executed based on the inherited age concept. Also, we propose a new sharing algorithm, called objective classication sharing algorithm(OCSA) that can preserve the diversity of the population. We will show the superior performance of the proposed algorithm by comparing the proposed algorithm with other promising algorithms for the test functions.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.8
• /
• pp.689-694
• /
• 2001

Recently, several promising multiobjective evolutionary algorithm such as SPEA. NSGA-II, PESA, and SPEA2 have been developed. In this paper, we also propose a new multiobjective evolutionary algorithm that compares to them. In the algorithm proposed in this paper, we introduce a novel concept, “inherited age” and total algorithm is executed based on the inherited age concept. Also, we propose a new sharing algorithm, called objective classication sharing algorithm(OCSA) that can preserve the diversity of the population. We will show the superior performance of the proposed algorithm by comparing the proposed algorithm with other promising algorithms for the test functions.