• Journal of the Korean Institute of Intelligent Systems
• /
• v.9 no.6
• /
• pp.650-656
• /
• 1999

In this paper, we study on the convergence of the evolution strategies by introducing the Lamarckian evolution and the Baldwin effect, and propose a random local searching and a reinforcement local searching methods. In the random local searching method some neighbors generated randomly from each individual are med without any other information, but in the reinforcement local searching method the previous results of the local search are reflected on the current local search. From the viewpoint of the purpose of the local search it is suitable that we try all the neighbors of the best individual and then search the neighbors of the best one of them repeatedly. Since the reinforcement local searching method based on the Lamarckian evolution and Baldwin effect does not search neighbors randomly, but searches the neighbors in the direction of the better fitness, it has advantages of fast convergence and an improvement on the global searching capability. In other words the performance of the evolution strategies is improved by introducing the learning, reinforcement local search, into the evolution. We study on the learning effect on evolution strategies by applying the proposed method to various function optimization problems.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.4 s.36
• /
• pp.93-102
• /
• 2005

Excessive traffic of P2P applications in the limited communication environment is considered as a network bandwidth problem. Moreover, Though P2P systems search a resource in the phase of search using weakly connected systems(peers' connection to P2P overlay network is very weakly connected), it is not guaranteed to download the very peer's resource in the phase of download. In previous P2P search algorithm (1), we had adopted the heuristic peer selection method based on Random Walks to resolve this problems. In this paper, we suggested an adaptive P2P search algorithm based on the previous algorithm(1) to reform the node distribution rate which is affected in unit peer ability. Also, we have adapted the discriminative replication method based on a query ratio to reduce traffic amount additionally. In the performance estimation result of this suggested system, our system works on a appropriate point of compromise in due consideration of the direction of searching and distribution of traffic occurrence.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.10
• /
• pp.5132-5148
• /
• 2017

Cyber attacks are evolving commensurate with recent developments in information security technology. Intrusion detection systems collect various types of data from computers and networks to detect security threats and analyze the attack information. The large amount of data examined make the large number of computations and low detection rates problematic. Feature selection is expected to improve the classification performance and provide faster and more cost-effective results. Despite the various feature selection studies conducted for intrusion detection systems, it is difficult to automate feature selection because it is based on the knowledge of security experts. This paper proposes a feature selection technique to overcome the performance problems of intrusion detection systems. Focusing on feature selection, the first phase of the proposed system aims at constructing a feature subset using a sequential forward floating search (SFFS) to downsize the dimension of the variables. The second phase constructs a classification model with the selected feature subset using a random forest classifier (RFC) and evaluates the classification accuracy. Experiments were conducted with the NSL-KDD dataset using SFFS-RF, and the results indicated that feature selection techniques are a necessary preprocessing step to improve the overall system performance in systems that handle large datasets. They also verified that SFFS-RF could be used for data classification. In conclusion, SFFS-RF could be the key to improving the classification model performance in machine learning.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.11
• /
• pp.2298-2304
• /
• 2015

In this paper, we provide a new approach which optimizes the vertical tilting angle of the base station for multi-cell multiple-input single-output (MISO) downlink active antenna systems (AAS). Instead of the conventional optimal algorithm which requires an exhaustive search, we propose simple and near optimal algorithms. First, we represent a large system approximation based vertical beamforming algorithm which is applied to the average sum rate by using the random matrix theory. Next, we suggest a signal-to-leakage-and-noise ratio (SLNR) based vertical beamforming algorithm which simplifies the optimization problem considerably. In the simulation results, we demonstrate that the performance of the proposed algorithms is near close to the exhaustive search algorithm with substantially reduced complexity.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.5 no.3
• /
• pp.83-91
• /
• 1995

A neural network with one or more layers of hidden units can be trained using the well-known error back propagation algorithm. According to this algorithm, the synaptic weights of the network are updated during the training by propagating back the error between the expected output and the output provided by the network. However, the error back propagation algorithm is characterized by slow convergence and the time required for training and, in some situation, can be trapped in local minima. A theoretical formulation of a new fast learning method based on tabu search method is presented in this paper. In contrast to the conventional back propagation algorithm which is based solely on the modification of connecting weights of the network by trial and error, the present method involves the calculation of the optimum weights of neural network. The effectiveness and versatility of the present method are verified by the XOR problem. The present method excels in accuracy compared to that of the conventional method of fixed values.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.14 no.5
• /
• pp.499-505
• /
• 2003

In the design of mobile wireless communication system, base station location is one of the most important parameters. Designing base station location, the cost must be minimized by combining various, complex parameters. We can solve this problem by combining optimization algorithm, such as Simulated Annealing, Tabu Search, Genetic Algorithm, Random Walk Algorithm that have been used extensively fur global optimization. This paper shows the 4 kinds of algorithm to be applied to the optimization of base station location for communication system and then compares, analyzes the results and shows optimization process of algorithm.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.6
• /
• pp.1861-1866
• /
• 2000

This paper describes a multi-constrained university timetabling problem that is a one of the field of artificial intelligent research area. For this problem, we propose the 2type edge graph that is can be represented time-conflict and day-conflict constraints simultaneously. The genetic algorithms are devised and considered for it. And we describe a method of local search in traditional random operator for its search efficiency. In computational experiments, the solutions of proposed method are average 71% costs that ware compared with solutions of random method in 10,000 iterations.

• IE interfaces
• /
• v.17 no.spc
• /
• pp.103-110
• /
• 2004

This paper presents a two-phase method for the vehicle routing problems with time windows(VRPTW). In a supply chain management(SCM) environment, timely distribution is very important problem faced by most industries. The VRPTW is associated with SCM for each customer to be constrained the time of service. In the VRPTW, the objective is to design the least total travel time routes for a fleet of identical capacitated vehicles to service geographically scattered customers with pre-specified service time windows. The proposed approach is based on ant colony optimization(ACO) and improvement heuristic. In the first phase, an insertion based ACO is introduced for the route construction and its solutions is improved by an iterative random local search in the second phase. Experimental results show that the proposed two-phase method obtains very good solutions with respect to total travel time minimization.

• Journal of Korean Institute of Industrial Engineers
• /
• v.30 no.2
• /
• pp.120-129
• /
• 2004

This paper provides a simple but effective method, cost relaxation to escape from a local optimum of the traveling salesman problem. We would find a better solution if we repeat a local search heuristic at a different initial solution. To find a different initial solution, we use the cost relaxation method relaxing the cost of arcs. We used the Lin-Kernighan algorithm as a local search heuristic. In experimental result, we tested large instances, 30 random instances and 34 real world instances. In real-world instances, we found average 0.17% better above the optimum solution than the Concorde known as the chained Lin-Kernighan. In clustered random instances, we found average 0.9% better above the optimum solution than the Concorde.

• Proceedings of the KSME Conference
• /
• 2001.11a
• /
• pp.698-702
• /
• 2001

This paper presents a combinatorial method to compute the solutions of optimization problem. The present hybrid algorithm is the synthesis of an artificial life algorithm and the random tabu search method. The hybrid algorithm is not only faster than the conventional artificial life algorithm, but also gives a more accurate solution. In addition, this algorithm can find all global optimum solutions. And the enhanced artificial life algorithm is applied to optimum design of high-speed, short journal bearings and the usefuless is verified through this example.