• The Transactions of the Korea Information Processing Society
• /
• v.4 no.11
• /
• pp.2769-2779
• /
• 1997

This Paper Proposes an efficient method for improving the training performance of the neural networks using a hybrid of conjugate gradient backpropagation algorithm and dynamic tunneling backpropagation algorithm The conjugate gradient backpropagation algorithm, which is the fast gradient algorithm, is applied for high speed optimization. The dynamic tunneling backpropagation algorithm, which is the deterministic method with tunneling phenomenon, is applied for global optimization. Conversing to the local minima by using the conjugate gradient backpropagation algorithm, the new initial point for escaping the local minima is estimated by dynamic tunneling backpropagation algorithm. The proposed method has been applied to the parity check and the pattern classification. The simulation results show that the performance of proposed method is superior to those of gradient descent backpropagtion algorithm and a hybrid of gradient descent and dynamic tunneling backpropagation algorithm, and the new algorithm converges more often to the global minima than gradient descent backpropagation algorithm.

• Journal of the Korean Geosynthetics Society
• /
• v.19 no.3
• /
• pp.9-21
• /
• 2020

In this study, the design method of the combined sheet pile was considered in the coastal landfill where sandy and clayey soils are mixed, and the behavior in excavation was analyzed. It was confirmed from the elasto-plastic analysis that the predicted behavior of the temporary facilities of earth retaining differs according to the type of the combined sheet pile method (Built up, Interlocking, Welding) and the analysis method (soldier pile method, continuous wall method). In the case of sheet pile member force, the results of the continuous wall analysis method predicted the most conservative results. When the stress ratio (calculation/allowance) of each member was analyzed based on the maximum member force of the combined sheet pile method, the maximum value was obtained for bending moment in the side pile and combined stress in the case of the strut. As a result of finite element analysis, the member force of the side pile was the largest in the short-term effective stress analysis condition, while the compressive force of the strut was large in the consolidation analysis. When comparing the results of the elasto-plastic analysis and the finite element analysis, the shear force of the side pile and the axial force of the strut were greatly evaluated in the elasto-plastic analysis, and the bending moment of the side pile was the largest in the short-term effective stress condition of the finite element analysis. In addition, the displacement of the side pile was predicted to be greater in the finite element analysis than in the elasto-plastic analysis.

• Economic and Environmental Geology
• /
• v.50 no.1
• /
• pp.15-26
• /
• 2017

Advanced argillic, argillic, and phyllic zones are the most important alteration patterns to predict the hidden ore body during exploration of hydrothermal deposits. We examined the quantitative relationship between the spectral absorption characteristics and the mineral content of the synthetic mixtures such as alunite-kaolinite and illite-kaolinite using short wavelength infrared (SWIR) spectroscopy. In the alunite-kaolinite mixtures, the spectral absorption characteristics of the alunite was highly correlated with the Hull quotient reflectance(0.99) and the kaolinite had the highest correlation with the Gaussian peak(0.92). Illite-kaolinite mixtures are essential for Gaussian deconvolution because of the overlap of absorption region. Illite and kaolinite mixtures indicate the high correlation of 0.93 and 0.98, respectively. The error ranges in the alunite-kaolinite(8%) and illite-kaolinite mixtures(5%) derived from SWIR were smaller than the ones(29% and 26%) obtained from X-ray diffraction(Rietveld) analysis. These results show that SWIR spectroscopic analysis is more reliable than XRD Rietveld analysis in terms of quantification of allowed minerals.

• Journal of Korean Library and Information Science Society
• /
• v.36 no.4
• /
• pp.133-153
• /
• 2005

This study is to analyze and suggest the KDC's facet structure. KDC is using the Decimal Classification's facet structure. New suggestions are : Clearly use standard subdivisions and add instruction like 'Add to base number $\~$ notation $\~$ from table $\~$' : auxiliary tables contents are need to revise and expand to make a add table found in schedules and provide numbers to be added to designated schedules numbers are necessary.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.7 no.4
• /
• pp.3-10
• /
• 1997

본 논문은 기존 비밀분산방식의 악세스 구조를 조합디자인 이론의 관점에서 해석함으로써 새로운 비밀분산방식이 구성될 수 있음을 보인다. 종래의 비밀분산방식으로서는 다항식보간을 이용하는 방법, 사영기하를 이용하는 방법등이 알려져 있으나 본 논문에서는 OA(orthogonal Arrary), t-(v,k l)디자인, 그룹분할 가능한 GD(Group Divisible)디자인이 갖는 행렬구조로부터 비밀분산방식의 악세스 구조를 정합시킴으로써 비밀분산방식을 새롭게 구성하고 있다. 이와같이 구성된 비밀분산방식은 기존 방식의 비밀 사이즈가 소수의 멱승 q에 의존하고 있는 반면 본 방식의 경우 조합디자인 파라메터에 관계하고 있으므로 비밀 사이즈 선택의 융통성이 있고 잘 알려진 조합적 구조를 이용함으로써 실현이 용이한 특징을 갖는다.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.37 no.2
• /
• pp.213-223
• /
• 2013

This study presents the process of fashion item development with velvet through creative thinking methods. Creativity is one of the most important requirements for a successful job career and education enhancing creative thinking is needed in the area of fabrication, product design, and marketing strategy development. Velvet was selected as a research stimulus because it is a luxurious fabric with various differential properties such as a soft touch, unique luster, excellent drapability, and fine physical properties. The research methodology included creative thinking methods review, the selection of the tools, idea sourcing and listing, sequential idea evaluation and sample product making. After review of the various creative thinking methods, a combination method and forced connection method were employed as research tools to confirm the usefulness of creative thinking training because of their independence of use and application simplicity. A total of 12 university students participated as subjects in this research. After some training, each student derived ten ideas for velvet products that utilized a combination method and forced connection method. A total of 120 ideas were evaluated for novelty, technical possibility, practicality, and marketability; subsequently, 24 ideas were adopted and developed as sample products. The effectiveness of creativity education in fabrication and product design classes was verified through the whole process of product planning.

• Proceedings of the KIEE Conference
• /
• 2008.10b
• /
• pp.489-490
• /
• 2008

본 고에서는 체내 심실신호를 농하여 신경학적 분석 및 다형성의 측면에서 심실세동이 일어나는 것을 예측하는 분석 알고리즘을 설계하였다. 신경학적 측면에서는 시계열 신호의 Peak to Peak Interval을 예측법과 0.15Hz를 기준으로 HRV 신호의 AR Burg 모델링을 통하여 고주파성과 저주파성을 나누어 교감신경과 부교감신경의 활동성 통한 신경학적 예측법을 제시하였으며 또한 체내 심실신호의 비선형적 특성을 고려한 Fractal Dimension을 생성시킴으로서 주기성의 특성과 다형성 통한 예측법을 제시하였다. 체내 심전도를 기반으로 Simulation 하였으며 각 분석별 조합을 통하여 최적의 예측 구조를 찾고자 하였다. 의학적 의미가 있는 민감도와 특이도를 판별하였으며 예측을 위한 수행시간을 실험하였다. 이를 통하여 자율신경 활성도와 다형성 판별을 조합한 방법이 심실세동 예측을 위한 민감도의 측면에서 가장 우수함을 나타내었고 시뮬레이션을 위만 시뮬레이터(Simulator) UI(User Interface)를 제시하였다.

• The Korean Journal of Applied Statistics
• /
• v.30 no.1
• /
• pp.181-193
• /
• 2017

The paper introduces bottom-up and optimal combination methods that can analyze and forecast hierarchical time series. These methods allow forecasts at lower levels to be summed consistently to upper levels without any ad-hoc adjustment. They can also potentially improve forecast performance in comparison to independent forecasts. We forecast regional traffic accident counts as time series data in order to identify efficiency gains from hierarchical forecasting. We observe that bottom-up or optimal combination methods are superior to independent methods in terms of forecast accuracy.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2006.05a
• /
• pp.164-170
• /
• 2006

This paper proposes a RSM-based hybrid evolutionary algorithm (RHEA) which combines the merits of the popular programs such as genetic algorithm (GA), tabu search method, response surface methodology (RSM). This algorithm, for improving the convergent speed that is thought to be the demerit of genetic algorithm, uses response surface methodology and simplex method. The mutation of GA offers random variety to finding the optimum solution. In this study, however, systematic variety can be secured through the use of tabu list. Efficiency of this method has been proven by applying traditional test functions and comparing the results to GA. And it was also proved that the newly suggested algorithm is very effective to find the global optimum solution to minimize the weight for avoiding the resonance of fresh water tank that is placed in the rear of ship. According to the study, GA's convergent speed in initial stages is improved by using RSM method. An optimized solution is calculated without the evaluation of additional actual objective function. In a summary, it is concluded that RHEA is a very powerful global optimization algorithm from the view point of convergent speed and global search ability.

• Journal of Korean Society of Forest Science
• /
• v.86 no.1
• /
• pp.17-24
• /
• 1997

As the first step to estimate the potential natural vegetation distribution of the globe, the best spherical interpolation method was developed to the temperature and precipitation which have close relation to the distribution pattern of world natural vegetation. For developing the interpolation method, a named Light Climatic Dataset composed of 1,060 stations around the globe was randomly divided into halves of feeding side and target side. The discrepancy between the observed and estimated values at the target stations was compared with combinations of parameters and methods. The estimated values were calculated to each combination which is all-out, constant radius and constant station methods in the selection of the feeding stations, n square reciprocal and negative exponential functions in weighting function of distance between feeding stations and each target, and oval weighting in direction of the feeding stations from each target. As a result, it turned out that the spherical interpolation with negative exponential weighting function fed from the constant radius stations ovally weighed yields the best estimates both for temperature and for precipitation. The parameters for temperature are $30^{\circ}$ in constant radius, 0.78 in negative exponential function and 0.4 in oval weighting, and for precipitation are $30^{\circ}$, 0.53 and 0.4, respectively.