• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.969-974
• /
• 2020

Research into hyper parameter optimization (HPO) has recently revived with interest in models containing many hyper parameters, such as deep neural networks. In this paper, we introduce the most widely used HPO methods, such as grid search, random search, and Bayesian optimization, and investigate their characteristics through experiments. The MNIST data set is used to compare results in experiments to find the best method that can be used to achieve higher accuracy in a relatively short time simulation. The learning rate and weight decay have been chosen for this experiment because these are the commonly used parameters in this kind of experiment.

• Coupled systems mechanics
• /
• v.13 no.1
• /
• pp.1-19
• /
• 2024

The effectiveness of coupling with a bolted rim is assessed in this research using a newly designed optimization algorithm. The current study, which is provided here, evaluates 10 contemporary metaheuristic approaches for enhancing the coupling with bolted rim design problem. The algorithms used are particle swarm optimization (PSO), crow search algorithm (CSA), enhanced honeybee mating optimization (EHBMO), Harmony search algorithm (HSA), Krill heard algorithm (KHA), Pattern search algorithm (PSA), Charged system search algorithm (CSSA), Salp swarm algorithm (SSA), Big bang big crunch optimization (B-BBBCO), Gradient based Algorithm (GBA). The contribution of the paper isto optimize the coupling with bolted rim problem by comparing these 10 algorithms and to find which algorithm gives the best optimized result. These algorithm's performance is evaluated statistically and subjectively.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.10a
• /
• pp.504-505
• /
• 2017

Existing morpheme analyzers classify the words used in writing documents. A system for extracting sentences and paragraphs based on a morpheme analyzer is being developed. However, there are very few systems that compress documents and extract important paragraphs. The algorithm proposed in this paper calculates the weights of the keyword written in the document and extracts the paragraphs containing the keyword. Users can reduce the time to understand the document by reading the paragraphs containing the keyword without reading the entire document. In addition, since the number of extracted paragraphs differs according to the number of keyword used in the search, the user can search various patterns compared to the existing system.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.30 no.1
• /
• pp.187-198
• /
• 2005

The structure of a Web site can prevent the search robots or crawling agents from confusion in the midst of huge forest of the Web pages. We formalize the view on the World Wide Web and generalize it as a hierarchy of Web objects such as the Web as a set of Web sites, and a Web site as a directed graph with Web nodes and Web edges. Our approach results in the optimal hierarchical structure that can maximize the weight, tf-idf (term frequency and inverse document frequency), that is one of the most widely accepted content centric measures in the information retrieval community, so that the measure can be used to embody the semantics of search query. The experimental results represent that the optimization model is an effective alternative in the dynamically changing Web environment by replacing conventional heuristic approaches.

• Journal of information and communication convergence engineering
• /
• v.21 no.1
• /
• pp.62-67
• /
• 2023

This paper proposes an imputation method using a bidirectional k-nearest components search based local linear regression method. The bidirectional k-nearest-components search method selects components in the dynamic range from the missing points. Unlike the existing methods, which use a fixed-size window, the proposed method can flexibly select adjacent components in an imputation problem. The weight values assigned to the components around the missing points are calculated using local linear regression. The local linear regression method is free from the rank problem in a matrix of dependent variables. In addition, it can calculate the weight values that reflect the data flow in a specific environment, such as a blackout. The original missing values were estimated from a linear combination of the components and their weights. Finally, the estimated value imputes the missing values. In the experimental results, the proposed method outperformed the existing methods when the error between the original data and imputation data was measured using MAE and RMSE.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.05a
• /
• pp.463-466
• /
• 2008

취업 정보 시스템에서의 인재정보 및 채용정보의 검색결과에서 원하는 일자리와 인재를 빠른 시간내에 검색하는 문제는 사용자 만족도에 밀접한 영향을 미치는 중요한 요소 중의 하나라고 할 것이다. 본 연구에서는 키워드를 이용하여 채용정보를 검색할 경우 채용정보 항목에 가중치 값을 적용했을 때의 정확도를 판단하기 위한 정량적 평가의 효과성을 검토한다. 채용정보 항목에 가중치 값을 적용했을 경우와 적용하지 않았을 경우의 채용검색 결과를 취업정보 시스템 관련 담당자를 대상으로 관련성 점수를 5점 척도로 평가한다. 실험결과, 채용정보 항목에 가중치를 적용하였을 때 관련 정확도 평가값이 더 높았으며, 상위 20개의 검색결과 평가값을 순위에 따라 정확성 추이를 살펴본 결과 순위가 높을수록 더 높은 정확성을 보였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.1185-1188
• /
• 1996

In this paper we present a genetic approach for trajectory control algorithm of balancing weight for IWR biped walking robot. The biped walking robot, IWR that was made by Automatic Control Lab. of Inha University has a trunk which stabilizes its walking by generating compensation moment. Trunk is composed of a revolute and a prismatic joint which roles balancing weight. The motion of balancing weight is determined by the gait of legs and represented by two linear second order ordinary differential equations. The solution of this equation must satisfy some constraints simultaneously to have a physical meaning. Genetic algorithm search for this feasible motion of balancing weight under some constraints. Simulation results show that feasible motion of balancing weight can be obtained by genetic algorithm.

• Nuclear Engineering and Technology
• /
• v.42 no.3
• /
• pp.271-278
• /
• 2010

While the deterministic lattice physics/depletion codes use leakage-corrected critical spectrum (although approximate due to the B1 buckling search employed), Monte Carlo depletion codes currently in use do not have such a feature in spite of their heterogeneity and continuous-energy modeling capability. This paper describes an approach to Monte Carlo depletion with leakage-corrected critical spectrum derived from first principles. This is based on the concept of albedo eigenvalue treated as weight of the reflected neutron in Monte Carlo simulation.

• Journal of Korean Physical Therapy Science
• /
• v.15 no.3
• /
• pp.73-79
• /
• 2008

Background : Treadmill training has been proposed as a useful adjunct to conventional physical therapy to restore ability to walk after stroke. The purpose of this study to inform clinical practise by evaluating the research evidence for the effectiveness of treadmill training after stroke. Methods : We searched to the effectiveness of any form of intervention for effect of treadmill training by Dankook University electronic library databases of Medline, Embase, Cinahl, Amed and PEDro combined with a hand search of papers published in relevant peer-reviewed journals. Any type of study relevant to the topic published in English during time period from 1980 to 2007 was included. Results : The literature search identified 35 studies. The included studies enrolled a total of 374 subjects. 1. Treadmill retraining without partial body weight support might be more effective than no treatment at improving gait velocity, get up and go time, gait endurance and step length symmetry. 2. Treadmill retraining with partial body weight support might be more effective than no treatment in improving step length symmetry, gait velocity, gait endurance and balance. 3. Treadmill retraining might be more effective with partial body weight support than without it at improving gait velocity and motor improvement as measured by The Stroke Rehabilitation Assessment of Movement(STREAM). 4. Treadmill retraining without partial body weight support may be no different from physiotherapy and increase gait velocity to conventional gait therapy. 5. Treadmill retraining with partial body weight support may be no different from physiotherapy for gait velocity, motor recovery and balance. Conclusion The review suggests that although treadmill training of gait, especially with partial body weight support, might Improve gait parameters and functional mobility, unless treadmill training is directed at Improving gait speed it might be no more effective than conventional physical therapy at improving gait parameters.

• Animal Bioscience
• /
• v.37 no.5
• /
• pp.832-838
• /
• 2024

Objective: The use of molecular genetic methods in pig breeding can significantly increase the efficiency of breeding and breeding work. We applied the F_st (fixsacion index) method, the main focus of the work was on the search for common options related to the number of born piglets and the weight of born piglets, since today the urgent task is to prevent a decrease in the weight of piglets at birth while maintaining high fertility of sows. Methods: One approach is to scan the genome, followed by an assessment of Fst and identification of selectively selected regions. We chose Large White sows (n = 237) with the same conditions of keeping and feeding. The data were collected from the sows across three farrowing. For genotyping, we used GeneSeek GGP Porcine HD Genomic Profiler v1, which included 68,516 single nucleotide polymorphisms evenly distributed with an average spacing of 25 kb (Illumina Inc, San Diego, CA, USA). Results: Based on the results of the Fst analysis, 724 variants representing selection signals for the signs BALWT, BALWT1, NBA, and TNB (weight of piglets born alive, average weight of the 1st piglets born alive, total number born alive, total number born). At the same time, 18 common variants have been identified that are potential markers for both the number of piglets at birth and the weight of piglets at birth, which is extremely important for breeding work to improve reproductive characteristics in sows. Conclusion: Our work resulted in identification of variants associated with the reproductive characteristics of pigs. Moreover, we identified, variants which are potential markers for both the number of piglets at birth and the weight of piglets at birth, which is extremely important for breeding work to improve reproductive performance in sows.