• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.241-265
• /
• 2023

A corporate insolvency prediction model serves as a vital tool for objectively monitoring the financial condition of companies. It enables timely warnings, facilitates responsive actions, and supports the formulation of effective management strategies to mitigate bankruptcy risks and enhance performance. Investors and financial institutions utilize default prediction models to minimize financial losses. As the interest in utilizing artificial intelligence (AI) technology for corporate insolvency prediction grows, extensive research has been conducted in this domain. However, there is an increasing demand for explainable AI models in corporate insolvency prediction, emphasizing interpretability and reliability. The SHAP (SHapley Additive exPlanations) technique has gained significant popularity and has demonstrated strong performance in various applications. Nonetheless, it has limitations such as computational cost, processing time, and scalability concerns based on the number of variables. This study introduces a novel approach to variable selection that reduces the number of variables by averaging SHAP values from bootstrapped data subsets instead of using the entire dataset. This technique aims to improve computational efficiency while maintaining excellent predictive performance. To obtain classification results, we aim to train random forest, XGBoost, and C5.0 models using carefully selected variables with high interpretability. The classification accuracy of the ensemble model, generated through soft voting as the goal of high-performance model design, is compared with the individual models. The study leverages data from 1,698 Korean light industrial companies and employs bootstrapping to create distinct data groups. Logistic Regression is employed to calculate SHAP values for each data group, and their averages are computed to derive the final SHAP values. The proposed model enhances interpretability and aims to achieve superior predictive performance.

• THE INTERNATIONAL COMMERCE & LAW REVIEW
• /
• v.25
• /
• pp.3-40
• /
• 2005

Many Korean construction companies have been performing a variety of overseas construction projects since the 1970s. It is unfortunate that in many cases they have had to suffer big losses caused by errors and defects in the design and construction of the projects. In the author's opinion, however, there were losses that could have been avoided if they had understood better the feature and content of the particular construction contract. Few lawyers and scholars in Korea have been interested in the research and study of international construction contracts. This is mainly because they do not have access to practical sources outside of Korea for them to research and study since the contracts undertaken have been dealt with by law firms in other jurisdictions to which the disputes apply. This article is aiming primarily at the introduction of the issues which the practitioners are likely to confront in the process of reviewing and performing the international construction contract. In some cases solutions are sought about these issues based upon the FIDIC standard terms and conditions, the actual experience of practice, and UNIDROIT Principles, etc. It is reasonable to say that all the issues related to the international construction contract cannot be covered in a short article like this. The author wishes this article could induce subsequent studies on international construction contracts for further research. It has to be noted that from time to time Korean construction practices have been compared to the international ones for better understanding. This article mostly includes cases where the Korean construction companies go overseas for their projects, while there are some cases where foreign developers and financial investors participated in domestic projects in which international construction contracts forms were adopted. A few precedent domestic writings about international construction contracts seems to lack emphasis on the points that there are several standard construction contract forms and that they are different. The differences are mainly in accordance with who bears the design responsibility, how the owner has to make progress payments to the contractor and who the funding source for the project is. This article tries to make it clear that there are significant differences between the standard contract forms, e.g. a simple construction form, a design-build form and an EPC/turnkey form of contract. Again, the author hopes that this article can arouse the interest in the international construction contracts from both academic and practical fields, so that many subsequent advanced articles can help our construction industry become much more competitive in the world through awareness of the methods of procurement and administration of the contracts.

• Proceedings of the Korean Institute Of Construction Engineering and Management
• /
• autumn
• /
• pp.353-358
• /
• 2001

Since the financial crisis broke out, liquidity has become the critical issue in housing construction industry. In order to secure liquidity, it is prerequisite to precisely forecast cash flow. However, construction companies have failed to come up with a systematic process to manage and forecast cash flow. Until now, companies have solely relied on the prediction of profits and losses, which is carried out as they review business feasibility. To obtain more accurate cash flow forecast model, practical pattern of payments should be taken into account. In this theory, basic model that analyzes practical housing payment collection pattern resulting from prepayments and arrears is described. This model is to complement conventional cash flow forecast scheme in the phase of business feasibility review. Analysis result on final losses in cash that occur as a result of prepayment and arrears is considered in this model. Additionally, in the estimation of construction cost in the phase of business feasibility review, real construction prices instead of official prices are applied to enhance accuracy of cash outflow forecast. The proportion of payment made by a bill and changes in payment date caused by rescheduling of a bill are also factored in to estimate cash outflow. This model would contribute to achieving accurate cash flow forecast that better reflect real situation and to enhancing efficiency in capital management by giving a clear picture with regard to the demand and supply timing of capital.

• International Journal of Highway Engineering
• /
• v.11 no.1
• /
• pp.117-126
• /
• 2009

Total number of recorded earthquakes in Korea is more than 2,000 of which 48 were catastrophic. The impacts from infrastructure damage due to an earthquake to production facilities and lifelines may spread across boundaries of several regions via import-export relationships and can bring serious economic impact to other regions. The economic impacts from unscheduled events stem not only from the damage and direct losses, but also from the indirect losses during the recovery and reconstruction periods. To recover and reconstruct the facilities and lifelines damaged by unexpected events through investment or government financial aid, both the direct and the indirect economic impacts from an event need to be measured in regional and interregional contexts. Direct economic impact is the direct change of production and demand due to the disruption of production facilities and lifelines from an unexpected event, and indirect economic impact is the change in other sectors due to inter-industry relationships. The purpose of the paper is to analyze various economic impacts of an earthquake, especially impacts on transportation networks in Korea. We collected spatial and economic data from Korea, and analyzed and estimated final demand loss and commodity flows from the unscheduled event.

• 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.

• Journal of the Korean Society of Safety
• /
• v.34 no.1
• /
• pp.115-121
• /
• 2019

Background: Due to the diversification and advancement of research, researchers have become to deal with a variety of chemical and biological harmful materials in the laboratories of universities and research institutes and the risk has increased as well. Therefore, it is necessary to strengthen the social safety net for laboratory accidents by strengthening the compensation to the level comparable to that of Korean Workers' Compensation & Welfare Service, when the researchers become physically disabled by laboratory accidents. The purpose of this study is to secure researchers' health rights and to create a research environment where researchers can work with confidence by strengthening the compensation to the level comparable to that of Korean Workers' Compensation & Welfare Service. Method: We analyzed the laboratory accidents by year, injury type, severity of accident and disability grade with the 6 year data from 2011 to 2016, provided by Laboratory Safety Insurance. Based on the analysis result, we predicted the financial impact on Laboratory Safety Insurance if we introduce a compensation annuity by disability grade which is similar to Injury-Disease Compensation Annuity of Korean Workers' Compensation & Welfare Service. Result :As of 2011, the insured number of Laboratory Safety Insurance was approximately 700,000. The Average premium per insured was KRW 3,339 and there were 158 claims. Total claim amount was KRW 130 million, whereas the premium was about KRW 2.3 billion. The loss ratio was very low at 5.75%. If we introduce a compensation annuity by disability grade similar to Injury-Disease Compensation Annuity of Korean Workers' Compensation & Welfare Service, the expected benefit amount for 1 case of disability grade 1 would be KRW 1.6 billion, assuming 2% of interest rate. Given current premium, the loss ratio, the ratio of premium income to claim payment, is expected 41.4% in 2017 and 151.6% in 2026. The increased loss ratio due to the introduce of the compensation annuity by disability grade is estimated to be 11.0% in 2017 and 40.4% in 2026. Conclusion: Currently, laboratories can purchase insurance companies' laboratory safety insurance that meets the standards prescribed by Act on the Establishment of Safe Laboratory Environment. However, if a compensation annuity is introduced, it would be difficult for insurance companies to operate the laboratory safety insurance due to financial losses from a large-scale accident. Therefore, it is desirable that one or designated entities operate laboratory safety insurance. We think that it is more desirable for laboratory safety insurance to be operated by a public entity rather than private entities.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.6 no.2
• /
• pp.471-483
• /
• 2014

Since 2009 of global financial crisis, shipbuilding industry has undergone hard times seriously. After such a long depression, the latest global shipping market index shows that the economic recovery of global shipbuilding market is underway. Especially, nations with enormous resources are going to increase their productivity or expanding their shipyards to accommodate a large amount of orders expected in the near future. However, few commercial projects have been carried out for the practical shipyard layout designs even though those can be good commercial opportunities for shipbuilding engineers. Shipbuilding starts with a shipyard construction with a large scale investment initially. Shipyard design and the equipment layout problem, which is directly linked to the productivity of ship production, is an important issue in the production planning of mass production of ships. In many cases, shipbuilding yard design has relied on the experience of the internal engineer, resulting in sporadic and poorly organized processes. Consequently, economic losses and the trial and error involved in such a design process are inevitable problems. The starting point of shipyard construction is to design a shipyard layout. Four kinds of engineering parts required for the shipyard layout design and construction. Those are civil engineering, building engineering, utility engineering and production layout engineering. Among these parts, production layout engineering is most important because its result is used as a foundation of the other engineering parts, and also, determines the shipyard capacity in the shipyard lifecycle. In this paper, the background of shipbuilding industry is explained in terms of engineering works for the recognition of the macro trend. Nextly, preliminary design methods and related case study is introduced briefly by referencing the previous research. Lastly, the designed work of layout design is validated using the computer simulation technology.

• THE INTERNATIONAL COMMERCE & LAW REVIEW
• /
• v.52
• /
• pp.119-135
• /
• 2011

Korea's trading volume is expected to surpass USD 1 trillion in 2011. Korean economy achieves this largely due to its dependence on export and enhanced technological capacity and product quality. Improved recognition of Korean enterprises in the global arena also helped. However, the largest reason behind theses could be found in exporters' ceaseless marketing endeavors and continued government supports. Today, more and more people become to rely on trade exhibitions to boost export effectively. Trade exhibitions are employed as a useful tool to attract buyers and enter a market. This is because such exhibitions' marketing effect and professional aspects. South Korean export relies on large conglomerates for most part. However, 95% of the country's entire industry are small and medium-sized companies. This means that SMEs' export has a huge impact on the national economy. Therefore, as a way to improve SME export and minimize their losses, we need to use trade exhibitions more actively. The overseas policy insures only promotional activities regarding foreign exhibition and fair (including international events held in Korea), foreign distributor, shopping mall, home shopping, etc. and does not include air fare, traffic cost in the local place, accommodation cost and dining expenditures as costs to be excluded mentioned above. It is not easy for them to take part in events abroad if they have to pay for such expenses. If full financial support is difficult, the Korea trade insurance corporation still may give a certain level of aid for successful exporters to engage in marketing activities abroad more actively and further stimulate SMEs' export.

• Journal of Digital Contents Society
• /
• v.19 no.6
• /
• pp.1185-1195
• /
• 2018

Industrial Control System (ICS) is a system that carry out monitoring and controls of industrial control process and is applied in infrastructure such as water, power, and gas. Recently, cyber attacks such as Brutal Kangaroo, Emotional Simian, and Stuxnet 3.0 have been continuously increasing in ICS, and these security risks cause damage of human life and massive financial losses. Attacks on the control layer among the attack methods for ICS can malfunction devices of the field device layer by manipulating control commands. Therefore, in this paper, we propose a mechanism that apply the SDN between the control layer and the field device layer in the industrial control system and to determine whether the control command is legitimate or not and we show simulation results on a simply composed control system.

• Journal of Convergence Society for SMB
• /
• v.5 no.3
• /
• pp.27-31
• /
• 2015

As the number of smartphone users is increasing with the development of mobile devices, the range of monetary transaction from the individual use is increasing. Therefore, hacking methods are diversified and the information forgery of mobile devices has been a current issue. The forgery via apps in mobile devices is a hacking method that creates an app similar to well-known apps to deceive the users. The forgery attack corresponds to the violation of integrity, one of three elements of security. Due to the forgery, the value and credibility of an app decreases with the risk increased. With the forgery in app, private information and data can be stolen and the financial losses can occur. This paper examined the forgery, and suggested a way to detect it, and sought the countermeasure to the forgery.