• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.83-102
• /
• 2021

The government recently announced various policies for developing big-data and artificial intelligence fields to provide a great opportunity to the public with respect to disclosure of high-quality data within public institutions. KSURE(Korea Trade Insurance Corporation) is a major public institution for financial policy in Korea, and thus the company is strongly committed to backing export companies with various systems. Nevertheless, there are still fewer cases of realized business model based on big-data analyses. In this situation, this paper aims to develop a new business model which can be applied to an ex-ante prediction for the likelihood of the insurance accident of credit guarantee. We utilize internal data from KSURE which supports export companies in Korea and apply machine learning models. Then, we conduct performance comparison among the predictive models including Logistic Regression, Random Forest, XGBoost, LightGBM, and DNN(Deep Neural Network). For decades, many researchers have tried to find better models which can help to predict bankruptcy since the ex-ante prediction is crucial for corporate managers, investors, creditors, and other stakeholders. The development of the prediction for financial distress or bankruptcy was originated from Smith(1930), Fitzpatrick(1932), or Merwin(1942). One of the most famous models is the Altman's Z-score model(Altman, 1968) which was based on the multiple discriminant analysis. This model is widely used in both research and practice by this time. The author suggests the score model that utilizes five key financial ratios to predict the probability of bankruptcy in the next two years. Ohlson(1980) introduces logit model to complement some limitations of previous models. Furthermore, Elmer and Borowski(1988) develop and examine a rule-based, automated system which conducts the financial analysis of savings and loans. Since the 1980s, researchers in Korea have started to examine analyses on the prediction of financial distress or bankruptcy. Kim(1987) analyzes financial ratios and develops the prediction model. Also, Han et al.(1995, 1996, 1997, 2003, 2005, 2006) construct the prediction model using various techniques including artificial neural network. Yang(1996) introduces multiple discriminant analysis and logit model. Besides, Kim and Kim(2001) utilize artificial neural network techniques for ex-ante prediction of insolvent enterprises. After that, many scholars have been trying to predict financial distress or bankruptcy more precisely based on diverse models such as Random Forest or SVM. One major distinction of our research from the previous research is that we focus on examining the predicted probability of default for each sample case, not only on investigating the classification accuracy of each model for the entire sample. Most predictive models in this paper show that the level of the accuracy of classification is about 70% based on the entire sample. To be specific, LightGBM model shows the highest accuracy of 71.1% and Logit model indicates the lowest accuracy of 69%. However, we confirm that there are open to multiple interpretations. In the context of the business, we have to put more emphasis on efforts to minimize type 2 error which causes more harmful operating losses for the guaranty company. Thus, we also compare the classification accuracy by splitting predicted probability of the default into ten equal intervals. When we examine the classification accuracy for each interval, Logit model has the highest accuracy of 100% for 0~10% of the predicted probability of the default, however, Logit model has a relatively lower accuracy of 61.5% for 90~100% of the predicted probability of the default. On the other hand, Random Forest, XGBoost, LightGBM, and DNN indicate more desirable results since they indicate a higher level of accuracy for both 0~10% and 90~100% of the predicted probability of the default but have a lower level of accuracy around 50% of the predicted probability of the default. When it comes to the distribution of samples for each predicted probability of the default, both LightGBM and XGBoost models have a relatively large number of samples for both 0~10% and 90~100% of the predicted probability of the default. Although Random Forest model has an advantage with regard to the perspective of classification accuracy with small number of cases, LightGBM or XGBoost could become a more desirable model since they classify large number of cases into the two extreme intervals of the predicted probability of the default, even allowing for their relatively low classification accuracy. Considering the importance of type 2 error and total prediction accuracy, XGBoost and DNN show superior performance. Next, Random Forest and LightGBM show good results, but logistic regression shows the worst performance. However, each predictive model has a comparative advantage in terms of various evaluation standards. For instance, Random Forest model shows almost 100% accuracy for samples which are expected to have a high level of the probability of default. Collectively, we can construct more comprehensive ensemble models which contain multiple classification machine learning models and conduct majority voting for maximizing its overall performance.

• The KIPS Transactions:PartC
• /
• v.10C no.3
• /
• pp.367-376
• /
• 2003

This study deals with the DCMST (Delay constrained Capacitated Minimum Spanning Tree) problem applied in the topological design of local networks or finding several communication paths from root node. While the traditional CMST problem has only the traffic capacity constraint served by a port of root node, the DCMST problem has the additional mean delay constraint of network. The DCMST problem consists of finding a set of spanning trees to link end-nodes to the root node satisfying the traffic requirements at end-nodes and the required mean delay of network. The objective function of problem is to minimize the total link cost. This paper presents two-phased heuristic algorithm, which consists of node exchange, and node shift algorithm based on the trade-off criterions, and mean delay algorithm. Actual computational experience and performance analysis show that the proposed algorithm can produce better solution than the existing algorithm for the CMST problem to consider the mean delay constraint in terms of cost.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.2 no.1
• /
• pp.38-60
• /
• 1999

The main objective of this study is systemization of the technique of ROC quantification and optimization of baseline design by applying CE principle to the acquisition process of a weapon system. QFD and TOA techniques can be employed to a good working example of the conceptual design of a future main battle tank. In this paper, Product Planning Phase, the first phase of four QFD phases, is deployed in terms of eight steps including customer requirements and final product control characteristics. TOA is carried out considering only combat weight. In order to perform combat weight analysis and performance TOA, Preliminary Configuration Synthesis Methodology is used. Preliminary Configuration Synthesis Methodology employs the method of least squares and described linear equations of weight interrelation equation for each component of tank. As a result of QFD based upon the ROC, it was cleared that armor piercing power, main armament, type of ammunition, cruising range, combat weight, armor protection, power loading, threat detection and cost are primary factors influencing design and that combat weight is the most dominant one. The results of TOA based on the combat weight constraint show that 5100 lb reduction was required to satisfy the ROC. The baseline design of a future main battle tank is illustrated with assumption that all phases of QFD are employed to development and production process of subsystems, components, and parts of main battle tank. TOA is applied in iterative process between initial baseline design and ROC. The detailed design of each component is illustrated for a future main battle tank.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.6
• /
• pp.1148-1156
• /
• 2015

The maritime cognitive radio networks (MCRNs) provide the high throughput with a low communication cost because the secondary ships opportunistically access to unused licensed bands of primary ships. In the ground cognitive radio networks, the busy and idle state of a primary user during a frame is not nearly changed because the state of the primary user are slowly changed. However, the state of the primary ship in the MCRNs may be frequently changed in the frame. Therefore, this paper evaluates the throughput of a primary ship and secondary ships in the MCRNs taking the state change of a primary ship into consideration when the fusion center uses the cooperative spectrum sensing. The simulation results show that trade-off between the throughput of a primary ship and that of secondary ships according to the system parameter such as the cooperative spectrum sensing scheme, the number of secondary ships, and the target detection probability.

• The Journal of Information Systems
• /
• v.25 no.3
• /
• pp.31-60
• /
• 2016

Purpose In 2016, the market size of mobile(m-) shopping goes beyeond more than half of a total of online shopping. People use smartphones as the main device for m-commerce. Under the circmustances, this study attempts to address why people prefer to use smartphone-based m-commerce. In other words, it is necessary to understand the main value that smartphone-based m-commerce creates. Drawing on the studies of consumption value, this study focuses on utilitarian value in predicting customers' continuance intention in the context of smartphone-based m-commerce, recognizing that utilitarian value is a key extrinsic motivation in the goal-oriented, performance-oriented shopping contexts. Furthermore, this study identifies factors affecting customers' utilitarian value from the perspective of benefits and costs, following the notion that it represents the result of evaluating a trade-off of benefits and costs caused by smartphone-based m commerce. More specifically, in this study, ubiquitous service, location-based service (LBS), transaction speed, and price utility belong to the benefit dimension, whereas technology anxiety and cognitive effort belong to the cost dimension. Design/methodology/approach To test the proposed hypotheses, the study conducted partial least squares (PLS) analysis with a total of 294 data collected on users with experience in smartphone-based m-commerce. Findings The results show that first, utilitarian value is increased by the benefits, such as ubiquitous service, transaction speed, and price utility. However, LBS has no direct effect on utilitarian value. Second, the noteworthy finding is that ubiquitous service and LBS greatly increase transaction speed. Third, technology anxiety and cognitive effort considered as the cost dimension are negatively associated with utilitarian value but their impacts on it are non-significant. Finally, the results support the argument that utilitarian value is a determinant of continuance intention. Overall, the findings imply that utilitarian value greatly depends on the peception on benefits rather than the aspect of cost in smartphone-based m-commerce. Overall, the findings offer new insight into the studies of m-commerce by considering and verifying the impacts of its benefits and costs simultaneously.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.10 no.4
• /
• pp.163-168
• /
• 2010

In this paper, we propose novel two-way relaying scheme in which number of two users transmit the signal to number of two destination. The scheme is difference with conventional multi-hop two-way relay scheme in which number of two users exchange the each signal. In the conventional scheme, each user have to perform back-propagating self-interference method to remove the own signal from received signal. It occurred to increase the complexity for signal processing at the user. To overcome the problem, we apply the Pre-cancellation which can reduce the process of back-propagating self-interference to our proposal network Simulation result show that proposal scheme outperform the convention multi-hop two-way relay scheme. Also we analysis the trad-off between performance and complexity accordance with using Pre-cancellation method.

• KDI Journal of Economic Policy
• /
• v.32 no.3
• /
• pp.71-99
• /
• 2010

The concept of CoVaR introduced by Adrian and Brunnermeier (2009) is a useful tool to measure the risk spillover effect. It can capture the risk contribution of each institution to overall systemic risk. While Adrian and Brunnermeier rely on the quantile regression method in the estimation of CoVaR, we propose a new estimation method using parametric distribution functions such as bivariate normal and $S_U$-normal distribution functions. Based on our estimates of CoVaR for Korean banking industry, we investigate the practical usefulness of CoVaR for a systemic risk measure, and compare the estimation performance of each model. Empirical results show that bank makes a positive contribution to system risk. We also find that quantile regression and normal distribution models tend to considerably underestimate the CoVaR (in absolute value) compared to $S_U$-normal distribution model, and this underestimation becomes serious when the crisis in a financial system is assumed.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.10
• /
• pp.175-185
• /
• 1998

In this paper, we studied some design methodologies of typical FIR filters based on the peak-error constrained least square criterion which was first introducedd by Adams in 1991. This method is a mixed type of the classical least squared error method(LSM) and the so-called min-max error method (MMM). And by considering both the least squared error as well as the maximum error, the solution, i.e. the impulse response of the filter, can be found only when the restrictions on maximum gain, transition bandwidth, and the squared error are satisfied simultaneously under some trade-off conditions. We used the multiple exchange algorithms for optimization procedure and applied the design methodology to the cases of the multiband filter, the differentiator, and the Hilbert transformer by taking the balance of two design criteria into account. The results show that the peak-error constrained least weighted square error design method(PLEM) is superior in performance to the existing LSM and MMM from both the squared error and the maximum error standpoints. And it is verified that PLEM can be applied to not only the case of simple low pass filter, but also to various types of FIR filters.

• Asia-Pacific Journal of Business
• /
• v.13 no.1
• /
• pp.167-183
• /
• 2022

Purpose - The purpose of this study is to show the need to use the past long-term returns for investment decisions in U.S. equity funds and to suggest an investment strategy using long-term returns. Design/methodology/approach - This study solves the problem of high return volatility in long-term returns and proposes new investment portfolios based on the behavior of fund investors according to past returns. For the investment portfolio of this study, 60 months are divided into several periods and the average of the performance ranks for each period is used. Findings - First, funds with high average returns over multiple periods have lower future outflows and higher future returns than funds with high 60-month cumulative returns. Second, funds with low average returns over multiple periods have lower future inflows and lower future returns than funds with low 60-month cumulative returns. The findings mean that when making decisions based on past long-term returns, it is a smarter investment choice to buy funds with high average returns over multiple periods and sell funds with low average returns over multiple periods. Research implications or Originality - This study shows that it is necessary to use long-term returns in fund investment by analyzing the characteristics of the portfolio based on past returns. In addition, the study is meaningful in that it suggests a way to use long-term returns more efficiently based on the behavior of fund investors and shows that such investments lead to higher returns in the future.

• Asia-Pacific Journal of Business
• /
• v.13 no.4
• /
• pp.91-107
• /
• 2022

Purpose - This study quantitatively examines the level of financial inclusion of a microfinance institution in each region and how this is changing recently, and examines the level of financial inclusion by region and various financial characteristic factors related to it. It was empirically verified what kind of significant impact actually has on the institution's major management performance indicators (stability, profitability, efficiency, and public interest). Design/methodology/approach - It was confirmed that the institution's financial inclusion index declined rapidly after 2015 as a whole, although there were some differences by region depending on regional characteristics. However, considering the fact that the number of branches per 100,000 adult population is steadily increasing nationwide, it was found that, contrary to what is known, the simple decrease in the number of branches of the institution was not the main cause. Findings - The analysis results of this study show that the institution's efforts for financial inclusion have a positive impact on profitability, stability, efficiency, and public interest, and that the institution pursues profitability, efficiency, stability, and public interest. showed that some trade-offs exist. In other words, overall, it was analyzed that profitability of the institution has a positive effect on efficiency, and efficiency has a positive effect on stability and public interest. Research implications or Originality - Since the institution's efforts to improve its profitability do not have a negative impact on its stability and public interest, it is judged that it is important to take a strategic stance, so excessive loan supply that exceeds the scope of the institution's own control needs to be avoided as much as possible. More detailed financial supply strategies and business management capabilities that enhance the asset soundness and management efficiency of safes need to be demonstrated.