• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 추계종합학술대회 논문집
• /
• pp.468-471
• /
• 1999

In this paper, we present a solution for combining multiple neural networks. Each neural network is trained with different features. And the neural networks are combined by four methods. The recognition rates by four combination methods are compared. The experimental results for handwritten digit recognition shows that the combination at hidden layers by single layer neural network is superior to any other methods. The reasons of the results are explained.

• 대한산업공학회지
• /
• 제38권3호
• /
• pp.198-207
• /
• 2012

In product or process parameter design, the case of multiple performance characteristics appears more commonly than that of a single characteristic. Numerous methods have been developed to deal with such multi-characteristic parameter design (MCPD) problems. Among these, this paper considers three representative methods, which are respectively based on the desirability function (DF), grey relational analysis (GRA), and principal component analysis (PCA). These three methods are then used to solve the MCPD problems in ten case studies reported in the literature. The performance of each method is evaluated for various combinations of its algorithmic parameters and alternatives. Relative performances of the three methods are then compared in terms of the significance of a design parameter and the overall performance value corresponding to the compromise optimal design condition identified by each method. Although no method is significantly inferior to others for the data sets considered, the GRA-based and PCA-based methods perform slightly better than the DF-based method. Besides, for the PCA-based method, the compromise optimal design condition depends much on which alternative is adopted while, for the GRA-based method, it is almost independent of the algorithmic parameter, and therefore, the difficulty involved in selecting an appropriate algorithmic parameter value can be alleviated.

• 한국산림과학회지
• /
• 제85권2호
• /
• pp.149-157
• /
• 1996

본 연구는 22개 지역에 대하여 임목수확 및 무육작업을 위한 임도개설순위를 결정하기 위하여, 다기준의사결정법중에 선호보정이 없는 모형인 최대최소법 및 최대최대법, 선호보정이 있는 모형인 단순가중치법, 계층가중치법 및 TOPSIS법의 적용성을 검토하였다. 그 결과 단순가중치법과 TOPSIS법이 임도개설순위의 결정에 적합한 것으로 판명되었다.

• 한국대기환경학회지
• /
• 제32권5호
• /
• pp.513-525
• /
• 2016

To propose an effective ensemble methods in predicting $PM_{10}$ concentration, six experiments were designed by different ensemble average methods (e.g., non-weighted, single weighted, and cluster weighted methods). The single weighted method was calculated the weighted value using both multiple regression analysis and singular value decomposition and the cluster weighted method was estimated the weighted value based on temperature, relative humidity, and wind component using multiple regression analysis. The effects of ensemble average methods were significantly better in weighted average than non-weight. The results of ensemble experiments using weighted average methods were distinguished according to methods calculating the weighted value. The single weighted average method using multiple regression analysis showed the highest accuracy for hourly $PM_{10}$ concentration, and the cluster weighted average method based on relative humidity showed the highest accuracy for daily mean $PM_{10}$ concentration. However, the result of ensemble spread analysis showed better reliability in the single weighted average method than the cluster weighted average method based on relative humidity. Thus, the single weighted average method was the most effective method in this study case.

• Journal of the Korean Statistical Society
• /
• 제31권2호
• /
• pp.163-175
• /
• 2002

An extended version of the minimax eccentricity factor estimation for multiple set case is proposed. In addition, two more simple methods for multiple set factor analysis exploiting the concept of generalized canonical correlation analysis is suggested. Finally, a certain connection between the generalized canonical correlation analysis and the multiple set factor analysis is derived which helps us clarify the relationship.

• Journal of applied mathematics & informatics
• /
• 제19권1_2호
• /
• pp.215-229
• /
• 2005

In this paper, constraint aggregation is combined with the adjoint and multiple shooting strategies for optimal control of differential algebraic equations (DAE) systems. The approach retains the inherent parallelism of the conventional multiple shooting method, while also being much more efficient for large scale problems. Constraint aggregation is employed to reduce the number of nonlinear continuity constraints in each multiple shooting interval, and its derivatives are computed by the adjoint DAE solver DASPKADJOINT together with ADIFOR and TAMC, the automatic differentiation software for forward and reverse mode, respectively. Numerical experiments demonstrate the effectiveness of the approach.

• 대한전자공학회논문지
• /
• 제26권12호
• /
• pp.2046-2054
• /
• 1989

Improved algorithms are presented to design linear phase digital FIR filters with coefficients of +1,0,-1 only followed by a multiple integrator. It has been shown that the existing linear phase filter design concept for the single integrator(or, accumulator)case can be extended to the case of the multiple integrator. Linear phase conditions for the multiple integrators are summarized. Filter design methods with double or triple integrator are exploited in datail and its computer simulation results are presented to deduce the advantages of multiple integrator to the single integrator.

• Journal of Korean Neurosurgical Society
• /
• 제53권2호
• /
• pp.102-107
• /
• 2013

Objective : The Leksell Gamma Knife$^{(R)}$ (LGK) is based on a single-fraction high dose treatment strategy. Therefore, independent verification of the Leksell GammaPlan$^{(R)}$ (LGP) is important for ensuring patient safety and minimizing the risk of treatment errors. Although several verification techniques have been previously developed and reported, no method has ever been tested statistically on multiple LGK target treatments. The purpose of this study was to perform and to evaluate the accuracy of a verification method (modified variable ellipsoid modeling technique, MVEMT) for multiple target treatments. Methods : A total of 500 locations in 10 consecutive patients with multiple brain tumor targets were included in this study. We compared the data from an LGP planning system and MVEMT in terms of dose at random points, maximal dose points, and target volumes. All data was analyzed by t-test and the Bland-Altman plot, which are statistical methods used to compare two different measurement techniques. Results : No statistical difference in dose at the 500 random points was observed between LGP and MVEMT. Differences in maximal dose ranged from -2.4% to 6.1%. An average distance of 1.6 mm between the maximal dose points was observed when comparing the two methods. Conclusion : Statistical analyses demonstrated that MVEMT was in excellent agreement with LGP when planning for radiosurgery involving multiple target treatments. MVEMT is a useful, independent tool for planning multiple target treatment that provides statistically identical data to that produced by LGP. Findings from the present study indicate that MVEMT can be used as a reference dose verification system for multiple tumors.

• 대한두개안면성형외과학회지
• /
• 제19권3호
• /
• pp.194-199
• /
• 2018

Background: Cases of simultaneous multiple skin cancers in a single patient have become more common. Due to the multiplicity of lesions, reconstruction in such cases is more difficult than after a single lesion is removed. This study presents a series of patients with multiple facial skin cancers, with an analysis of the surgical removal, reconstruction process, and the results observed during follow-up. Methods: We reviewed 12 patients diagnosed with multiple skin cancers on the face between November 2004 and March 2016. The patients' medical records were retrospectively reviewed to identify the type of skin cancer, the site of onset, methods of surgical removal and reconstruction, complications, and recurrence during follow-up. Results: Nine patients had a single type of cancer occurring as multiple lesions, while three patients had different skin cancer types that occurred together. A total of 30 cancer sites were observed in the 12 patients. The most common cancer site was the nose. Thirteen defects were reconstructed with a flap, while 18 were reconstructed with skin grafting. The only complication was one case of recurrence of basal cell carcinoma. Conclusion: Multiple skin cancers are removed by performing Mohs micrographic surgery or wide excision, resulting in multiple defect sites. The authors emphasize the importance of thoroughly evaluating local lesions surrounding the initially-identified lesions or on other sites when reconstructing a large defect which can not be covered by primary closure. Furthermore, satisfactory results can be obtained by using various methods simultaneously regarding the condition of individual patients, the defect site and size, and the surgeon's preference.

• Journal of the Korean Statistical Society
• /
• 제35권2호
• /
• pp.193-199
• /
• 2006

We consider a zero coupon bond that is at the risk of multitype multiple defaults. Assuming defaults occur according to k Cox processes, we find a price process for zero coupon bonds. To derive this process we follow the Lando (1998)'s method which uses conditional expectations instead of the traditional methods.