• Journal of Intelligence and Information Systems
• /
• v.9 no.1
• /
• pp.227-249
• /
• 2003

Prediction of corporate failure using past financial data is a well-documented topic. Early studies of bankruptcy prediction used statistical techniques such as multiple discriminant analysis, logit and probit. Recently, however, numerous studies have demonstrated that artificial intelligence such as neural networks can be an alternative methodology for classification problems to which traditional statistical methods have long been applied. In building neural network model, the selection of independent and dependent variables should be approached with great care and should be treated as model construction process. Irrespective of the efficiency of a teaming procedure in terms of convergence, generalization and stability, the ultimate performance of the estimator will depend on the relevance of the selected input variables and the quality of the data used. Approaches developed in statistical methods such as correlation analysis and stepwise selection method are often very useful. These methods, however, may not be the optimal ones for the development of neural network model. In this paper, we propose a genetic algorithms approach to find an optimal or near optimal input variables fur neural network modeling. The proposed approach is demonstrated by applications to bankruptcy prediction modeling. Our experimental results show that this approach increases overall classification accuracy rate significantly.

• Journal of KIISE:Software and Applications
• /
• v.28 no.9
• /
• pp.669-680
• /
• 2001

Genetic Algorithms are optimization algorithm that mimics biological evolution to solve optimization problems. Genetic algorithms provide an alternative to traditional optimization techniques by using directed random searches to locate optimal solutions in complex fitness landscapes. Hybrid genetic algorithm that is combined with local search called learning can sustain the balance between exploration and exploitation. The genetic traits that each individual in the population learns through evolution are transferred back to the next generation, and when this learning is combined with genetic algorithm we can expect the improvement of the search speed. This paper proposes a genetic algorithm based Cellular Learning with accelerated learning capability for function optimization. Proposed Cellular Learning strategy is based on periodic and convergent behaviors in cellular automata, and on the theory of transmitting to offspring the knowledge and experience that organisms acquire in their lifetime. We compared the search efficiency of Cellular Learning strategy with those of Lamarckian and Baldwin Effect in hybrid genetic algorithm. We showed that the local improvement by cellular learning could enhance the global performance higher by evaluating their performance through the experiment of various test bed functions and also showed that proposed learning strategy could find out the better global optima than conventional method.

• Korean Journal of Remote Sensing
• /
• v.27 no.1
• /
• pp.59-68
• /
• 2011

The analysis of optimal infiltration path is one of the representative fields in which the GIS technology can be useful for the military purpose. Usually the analysis of the optimal path is done with network data. However, for military purpose, it often needs to be done with raster data. Because raster data needs far more computation than network data, it is difficult to apply the methods usually used in network data, such as Dijkstra algorithm. The genetic algorithm, which has shown great outcomes in optimization problems, was applied. It was used to minimize the detection probability of infiltration route. 2D binary array genes and its crossover and mutation were suggested to solve this problem with raster data. 30 tests were performed for each population size, 500, 1000, 2000, and 3000. With each generation, more adoptable routes survived and made their children routes. Results indicate that as the generations increased, average detection probability decreased and the routes converged to the optimal path. Also, as the population size increases, more optimal routes were found. The suggested genetic algorithm successfully finds the optimal infiltration route, and it shows better performance with larger population.

• Journal of KIISE:Software and Applications
• /
• v.34 no.12
• /
• pp.1045-1055
• /
• 2007

Latent topic models are statistical models which automatically captures salient patterns or correlation among features underlying a data collection in a probabilistic way. They are gaining an increased popularity as an effective tool in the application of automatic semantic feature extraction from text corpus, multimedia data analysis including image data, and bioinformatics. Among the important issues for the effectiveness in the application of latent topic models to the massive data set is the efficient learning of the model. The paper proposes an accelerated learning technique for PLSA model, one of the popular latent topic models, by an incremental EM algorithm instead of conventional EM algorithm. The incremental EM algorithm can be characterized by the employment of a series of partial E-steps that are performed on the corresponding subsets of the entire data collection, unlike in the conventional EM algorithm where one batch E-step is done for the whole data set. By the replacement of a single batch E-M step with a series of partial E-steps and M-steps, the inference result for the previous data subset can be directly reflected to the next inference process, which can enhance the learning speed for the entire data set. The algorithm is advantageous also in that it is guaranteed to converge to a local maximum solution and can be easily implemented just with slight modification of the existing algorithm based on the conventional EM. We present the basic application of the incremental EM algorithm to the learning of PLSA and empirically evaluate the acceleration performance with several possible data partitioning methods for the practical application. The experimental results on a real-world news data set show that the proposed approach can accomplish a meaningful enhancement of the convergence rate in the learning of latent topic model. Additionally, we present an interesting result which supports a possible synergistic effect of the combination of incremental EM algorithm with parallel computing.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.4 no.3
• /
• pp.217-226
• /
• 2006

The structural and contamination characteristics of hot cells at KAERI were investigated. The SEM results showed that the size of the hot particulate on the inner surface of the hot cell ranged from 0.2 to $10{\mu}m$. It was found that an inlet flow rate of 15 m/sec was suitable for this developed cyclone with a 49 mm optimum vortex finder length. The results showed that the collection efficiency was about 85% for $3{\mu}m$ particles. The collection efficiency didn't show a sharp increase when the inlet flow rate was faster than 15m/sec. When the temperature of the inlet flow gas was increased, the collection efficiency of the cyclone was slightly decreased. The larger the vortex finder length was, the higher the pressure drop in the cyclone was. The cut size diameter decreased with an increment of the Reynolds number. It was established that the flow in the cyclone was a turbulent flow on the basis of the Reynolds number and this turbulent flow caused a pressure drop in the cyclone. $Stk^{1/2}_{50}$ decreased with increasing values of the Reynolds number and it gradually approached a constant value at a higher value of the Reynolds number Namely, $Stk^{1/2}_{50}$ approached approximately 0.045 between 6000 and 8000 of the Reynolds number.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.2
• /
• pp.697-703
• /
• 2017

This paper examines the flexure behavior of FRP-concrete composite structure that can replace conventional reinforced concrete structure types. In order to investigate the structural performance and behavioral characteristics in numerical analysis means, ABAQUS, a general purpose finite element analysis program, was utilized for nonlinear finite element analysis, and the various variables and their influences were analyzed and compared with experimental results to suggest values optimized to this composite structure. The concrete damage plasticity model and Euro code for concrete were used. In the implicit finite element analysis, the convergence was ambiguous when geometrical and material nonlinearity were large, so the explicit finite element analysis used in this study was deemed to be appropriate. From the comparison with the experiment about concrete damaged plasticity model, 20mm for the mesh size, $30^{\circ}$ for the dilation angle, $100Nmm/mm^2$ for the value of fracture energy, 0.667 for Kc value, and the consideration of damage parameter were suggested believed to be appropriate. The numerical model suggested in this study was able to imitate the ultimate load and cracking pattern very well; therefore, it is expected to be utilized in research of various new material composite structures.

• Korean Journal of Construction Engineering and Management
• /
• v.7 no.2 s.30
• /
• pp.98-108
• /
• 2006

The cost of construction engineering field is less than $3{\sim}5%$ of the gross construction cost, it, however, is very important because the performance of final facility depends on the technical ability of construction engineering company. However, as the results of the technical level of domestic construction engineering survey from experts, it was less than 70% of developed country. For this reason, domestic construction engineering companies are confronted with serious problems to go into overseas market. Accordingly, this research analyzed the construction engineering market trend of domestic and overseas in order to suggest the strategies to go into the Vietnam. Moreover, this research identified problems and strategies of making inroads into the world market through the survey from the domestic construction company related to overseas construction, advanced a theory through extensive literature review, and collected extensive opinions through an interview with domestic and foreign experts and business trip. The purpose of this research is to provide market analysis and strategic plan for advancing into overseas market in order to make the domestic construction engineering companies to go into the Vietnam market. Especially, as a plan for the revitalization of advancing overseas market of domestic construction engineering companies, this research proposed Constructing Local Infrastructure, Business Creation of Planning Proposals and Enlarged Plans of ODA(Official Development Assistant).

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.15 no.5
• /
• pp.95-107
• /
• 2016

The axle weight of a vehicle in motion can be measured with a low-speed or high-speed weigh-in-motion (WIM). However, the axial load dynamically change depending on the vehicle's characteristics-such as the chassis or axle structure-or the characteristics of the driving environment such as road flatness. The changes in dynamic load lead to differences between the vehicle's weight measured at rest and the vehicle's weight measured in motion. For this Study, an experiment was conducted with an instrumented vehicle to analyze the range of errors caused by uncontrollable environmental factors by identifying the characteristics of the dynamic load changes of a vehicle in motion, and determine the appropriate scale for the accuracy evaluation of a high-speed WIM, as a preparatory research for the introduction of unmanned overweight enforcement systems in the future. The key findings from the experiment are summarized as follows. First, The gross weight of the tested vehicle changed by approximately 1% at low velocities and approximately by 4% at high velocities, and the vehicle's axle weight changed by approximately 1-3%, at low velocities and by 2-9% at high velocities. A single axle showed larger weight changes than individual axles in a group. Secondly, The vehicle's gross weight and the axle weight on the impact section were up to eight times and three-to-twelve times higher, respectively, than its gross weight and the axle weight on the flat section. The vibration frequency of the vehicle's dynamic load was measured at between 2.4 and 5.8Hz, and found to return to the normal amplitude after moving approximately 30 meters.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.4
• /
• pp.21-29
• /
• 2011

At existing skin detection methods using skin color information defined based on the prior knowldege, threshold value to be used at the stage of dividing the backround and the skin region was decided on a subjective point of view through experiments. Also, threshold value was selected in a passive manner according to their background and illumination environments in these existing methods. These existing methods displayed a drawback in that their performance was fully influenced by the threshold value estimated through repetitive experiments. To overcome the drawback of existing methods, this paper propose a skin region detection method using a histogram approximation based on the mean shift algorithm. The proposed method is to divide the background region and the skin region by using the mean shift method at the histogram of the skin-map of the input image generated by the comparison of the similarity with the standard skin color at the CbCr color space and actively finding the maximum value converged by brightness level. Since the histogram has a form of discontinuous function accumulated according to the brightness value of the pixel, it gets approximated as a Gaussian Mixture Model (GMM) using the Bezier Curve method. Thus, the proposed method detects the skin region by using the mean shift method and actively finding the maximum value which eventually becomes the dividing point, not by using the manually selected threshold value unlike other existing methods. This method detects the skin region high performance effectively through experiments.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.17 no.3
• /
• pp.319-332
• /
• 2004

Due to their high strength to weight ratios and excellent durability, fiber reinforced polymer(FRP) is widely used in construction industries. In this paper, a shape optimum design of FRP bridge decks haying pultruded cellular cross-section is presented. In the problem formulation, an objective function is selected to minimize the volumes. The cross-sectional dimensions and material properties of the deck of FRP bridges are used as the design variables. On the other hand, deflection limits in the design code, material failure criteria, buckling load, minimum height, and stress are selected as the design constraints to enhance the structural performance of FRP decks. In order to efficiently treat the optimization process, the cross-sectional shape of bridge decks is assumed to be a tube shape. The optimization process utilizes an improved Genetic Algorithms incorporating indexing technique. For the structural analysis using a three-dimensional finite element, a commercial package(ABAQUS) is used. Using a computer program coded for this study, an example problem is solved and the results are presented with sensitivity analysis. The bridge consists of a deck width of 12.14m and is supported by five 40m long steel girders spaced at 2.5m. The bridge is designed to carry a standard DB-24 truck loading according to the Standard Specifications for Highway Bridges in Korea. Based on the optimum design, viable cross-sectional dimensions for FRP decks, suitable for pultrusion process are proposed.