• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.7 no.3
• /
• pp.518-523
• /
• 2006

Due to keen competition among companies, they have segmented customers and they are trying to offer specially targeted customer by means of the distinguished method. In accordance, data mining techniques are noted as the effective method that extracts useful information. This paper explores customer segmentation of the home study company using a hybrid decision tree and artificial neural network model. With the application of variance selection process from decision tree, the systemic process of defining input vector's value and the rule generation were developed. In point of customer management, this research analyzes current customers and produces the patterns of them so that the company can maintain good customer relationship. The case study shows that the predicted accuracy of the proposed model is higher than those of regression, decision tree (CART), artificial neural networks.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.16 no.4
• /
• pp.99-106
• /
• 2012

This study presents estimation method of temperature-dependent thermal conductivity by using solution of inverse heat conduction problem. Time and depth temperature distribution data from full-scale fire test were used for estimating temperature-dependent thermal conductivity on hybrid-fiber reinforced shield tunnel lining. At short heating time, estimated thermal conductivity sharply decreased within $100^{\circ}C$. On the other hand, it reflected thermal properties of concrete and effect of steel fiber at heating time of measured maximum heating temperature. Thus arbitrary time should be determined to estimate temperature-dependent thermal conductivity in time zone of measured maximum heating temperature. Estimated temperature-dependent thermal conductivity is similar to results of other study.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.6
• /
• pp.151-171
• /
• 2023

For the efficient establishment of a channel strategy for banks, this study aims to propose a channel model by classifying channels into types, and carrying out a correspondence analysis per type. A survey of bankers was conducted to visualize categorical data and create a positioning map. As a result, first, 12 banking channels were classified into 4 types based on business processing subjects and places, which were then, further grouped into the categories of full-banking and self-banking. Second, a correspondence analysis according to the classified types was carried out, and it was found that the branch-type is suitable for product description and customer management, while the banking-type is suitable for efficient business processing without time and space constraints. Furthermore, the analysis also showed that the machine-type and banking-type are inappropriate for customer management, and the mobility-type demonstrates low operational effectiveness due to a lack of awareness. The aforementioned findings suggest the need for a hybrid convergence channel that reflects the characteristics of banking tasks and fills in the gaps between the different channels. Third, a channel model was derived by adding a common area to the 2×2 model consisting of the business processing subjects and places. Therefore, this study is meaningful in that it examines the diversification of channels and factors in the division of roles by channel type based on customers' banking channel selection factors, and presents basic research findings for future channel strategy establishment and efficient channel operation.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.61-61
• /
• 2019

우리나라는 대부분이 산지(약 65%)로 구성되어 있어 강우 시 그 공간적 분포의 변동성이 매우 큰 편이며, 특히 전형적인 산지지형인 댐 유역의 경우 고도 변화 등에 기인한 지형특성 등에 따라 강우의 형태 및 패턴과 이에 따른 유출변화가 큰 복잡한 특성을 갖는다. 이로 인해 단순히 지점강우들을 공간보간(평균)한 면적강우를 홍수 유출모의 등에 활용할 경우 그 신뢰도가 매우 낮은 경우가 많아, 수문모의에 있어 레이더에 기반을 둔 공간 분포형 강우 등의 도입 검토가 요구된다. 한편, 최근 기상청에서는 보다 정확한 레이더 강수량 추정 값의 제공을 위해 "레이더-AWS 강우강도(Radar-AWS Rainrates, RAR)" 산출 기술을 지속적으로 개선하고 있으며, 이는 지상 우량계 대비 상당한 정확도를 보이고 있다. 본 연구에서는 국내 산지지형을 대표하며, 타 댐 유역에 비해 비교적 수문(수위/유량)관측소와 자료가 많은 용담시험유역에 기상레이더 강수량 추정 값(RAR)을 적용해 산지지형 댐 유역에서 강우의 시공간적 변동성과 이에 따른 홍수량의 정확한 분석을 통해 홍수 시 댐 유입량의 정확한 산정 등에 활용할 목적으로 홍수 유출모의를 수행하고자 한다. 모의에는 최근 5년(2014~2018년)동안 발생한 비교적 독립적인 1~2개(연도별)의 홍수사상을 적용하였으며, 모형은 분포형 강우를 적용할 수 있는 비교적 간단한 모형인 HEC-HMS를 활용하였다. HEC-HMS는 주로 집중형 수문모형(Lumped Hydrologic Model)으로 분류되어 레이더 강우와 같은 분포형 자료의 입력을 주로 적용치는 않고 있지만, HEC-GeoHMS와 ModClark 방법을 활용하면 격자단위의 분포형 강우를 적용할 수 있는 형태의 모델 구축이 가능하다. 모의 결과는 기존 유역평균 강우를 적용한 방법과 비교를 통해 그 개선점을 검토하고자 하며, 이를 통하여 산지지역 댐 유역의 홍수특성을 보다 더 정확하게 분석해보고자 한다. 한편, ModClark을 적용한 홍수 유출모의는 단순히 소유역별 도달시간의 격자별 비율을 고려한 홍수추적으로 그 해석상의 한계가 있어, 최근 개발된 하이브리드 수문모형(Hybrid Hydrologic Model, Distributed-Clark) 등도 동일유역에 대해 도입 적용할 계획에 있다.

• Journal of Intelligence and Information Systems
• /
• v.21 no.3
• /
• pp.79-99
• /
• 2015

The prediction of bankruptcy has been extensively studied in the accounting and finance field. It can have an important impact on lending decisions and the profitability of financial institutions in terms of risk management. Many researchers have focused on constructing a more robust bankruptcy prediction model. Early studies primarily used statistical techniques such as multiple discriminant analysis (MDA) and logit analysis for bankruptcy prediction. However, many studies have demonstrated that artificial intelligence (AI) approaches, such as artificial neural networks (ANN), decision trees, case-based reasoning (CBR), and support vector machine (SVM), have been outperforming statistical techniques since 1990s for business classification problems because statistical methods have some rigid assumptions in their application. In previous studies on corporate bankruptcy, many researchers have focused on developing a bankruptcy prediction model using financial ratios. However, there are few studies that suggest the specific types of bankruptcy. Previous bankruptcy prediction models have generally been interested in predicting whether or not firms will become bankrupt. Most of the studies on bankruptcy types have focused on reviewing the previous literature or performing a case study. Thus, this study develops a model using data mining techniques for predicting the specific types of bankruptcy as well as the occurrence of bankruptcy in Korean small- and medium-sized construction firms in terms of profitability, stability, and activity index. Thus, firms will be able to prevent it from occurring in advance. We propose a hybrid approach using two artificial neural networks (ANNs) for the prediction of bankruptcy types. The first is a back-propagation neural network (BPN) model using supervised learning for bankruptcy prediction and the second is a self-organizing map (SOM) model using unsupervised learning to classify bankruptcy data into several types. Based on the constructed model, we predict the bankruptcy of companies by applying the BPN model to a validation set that was not utilized in the development of the model. This allows for identifying the specific types of bankruptcy by using bankruptcy data predicted by the BPN model. We calculated the average of selected input variables through statistical test for each cluster to interpret characteristics of the derived clusters in the SOM model. Each cluster represents bankruptcy type classified through data of bankruptcy firms, and input variables indicate financial ratios in interpreting the meaning of each cluster. The experimental result shows that each of five bankruptcy types has different characteristics according to financial ratios. Type 1 (severe bankruptcy) has inferior financial statements except for EBITDA (earnings before interest, taxes, depreciation, and amortization) to sales based on the clustering results. Type 2 (lack of stability) has a low quick ratio, low stockholder's equity to total assets, and high total borrowings to total assets. Type 3 (lack of activity) has a slightly low total asset turnover and fixed asset turnover. Type 4 (lack of profitability) has low retained earnings to total assets and EBITDA to sales which represent the indices of profitability. Type 5 (recoverable bankruptcy) includes firms that have a relatively good financial condition as compared to other bankruptcy types even though they are bankrupt. Based on the findings, researchers and practitioners engaged in the credit evaluation field can obtain more useful information about the types of corporate bankruptcy. In this paper, we utilized the financial ratios of firms to classify bankruptcy types. It is important to select the input variables that correctly predict bankruptcy and meaningfully classify the type of bankruptcy. In a further study, we will include non-financial factors such as size, industry, and age of the firms. Thus, we can obtain realistic clustering results for bankruptcy types by combining qualitative factors and reflecting the domain knowledge of experts.

• Journal of the Korean Data and Information Science Society
• /
• v.12 no.2
• /
• pp.11-25
• /
• 2001

There are several algorithms for classification in modeling relations, patterns, and rules which exist in data. We learn to classify objects on the basis of instances presented to us, not by being given a set of classification rules. The Bayesian learning uses the probability distribution to express our knowledge about unknown parameters and update our knowledge by the law of probability as the evidence gathered from data. Also, the neural network models are designed for predicting an unknown category or quantity on the basis of known attributes by training. In this paper, we compare the misclassification error rates of Bayesian Neural Network method with those of other classification algorithms, CHAID, CART, and QUBST using several data sets.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2005.04a
• /
• pp.201-208
• /
• 2005

Neuro-Fuzzy modeling approach is proposed to predict the dynamic behavior of a single-degree-of-freedom structure that is equipped with hybrid base isolation system. Hybrid base isolation system consists of friction pendulum systems (FPS) and a magnetorheological (MR) damper. Fuzzy model of the M damper is trained by ANFIS using various displacement, velocity, and voltage combinations that are obtained from a series of performance tests. Modelling of the FPS is carried out with a nonlinear analytical equation that is derived in this study and neuro-fuzzy training. Fuzzy logic controller is employed to control the command voltage that is sent to MR damper. The dynamic responses or experimental structure subjected to various earthquake excitations are compared with numerically simulated results using neuro-fuzzy modeling method. Numerical simulation using neuro-fuzzy models of the MR damper and FPS predict response of the hybrid base isolation system very well.

• Journal of the Ergonomics Society of Korea
• /
• v.20 no.2
• /
• pp.29-46
• /
• 2001

This study is about the modeling of comfort feeling structure in indoor environments. To represent the degree of practical comfort feeling level in an environment, we measured elements of human sense and resultant elements of comfort feeling such as coziness, refreshment, and freshness with physical values(temperature, illumination, noise. etc.). The relationships of elements of human sense and elements of comfort feeling were formulated as a fuzzy model. And a hybrid-neural network with three layers were designed where obtained from fuzzy membership function values of the elements of human sense were used as inputs, and given as fuzzy membership function values of resultant elements of comfort feeling were used as outputs. Both kinds of fuzzy membership function values were obtained from physical values. The network was trained by measured data set. The proposed hybrid-neural network were tested and proposed a more realistic model of comfort feeling structure in indoor environments.

• The Korean Journal of Applied Statistics
• /
• v.27 no.3
• /
• pp.357-371
• /
• 2014

There are many studies related to imbalanced data in which the class distribution is highly skewed. To address the problem of imbalanced data, previous studies deal with resampling techniques which correct the skewness of the class distribution in each sampled subset by using under-sampling, over-sampling or hybrid-sampling such as SMOTE. Ensemble methods have also alleviated the problem of class imbalanced data. In this paper, we compare around a dozen algorithms that combine the ensemble methods and resampling techniques based on simulated data sets generated by the Backbone model, which can handle the imbalance rate. The results on various real imbalanced data sets are also presented to compare the effectiveness of algorithms. As a result, we highly recommend the resampling technique combining ensemble methods for imbalanced data in which the proportion of the minority class is less than 10%. We also find that each ensemble method has a well-matched sampling technique. The algorithms which combine bagging or random forest ensembles with random undersampling tend to perform well; however, the boosting ensemble appears to perform better with over-sampling. All ensemble methods combined with SMOTE outperform in most situations.

• Environmental and Resource Economics Review
• /
• v.21 no.2
• /
• pp.237-269
• /
• 2012

This study applies a discrete-continuous choice model to a national survey data set of automobile uses to investigate the potential impacts of a bonus-malus system for new cars in Korea. Not only the impacts on the discrete choice of automobile type and class but also those on the continuous decision making of car operation are analyzed. The characteristics of automobiles and individuals that determine car choice and operation are identified. The simulation based on the estimation result shows that an appropriately designed bonus-malus system can induce a reduction in energy consumption and $CO_2$ emission substantially without additional government expenditure.