• 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 Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.103-128
• /
• 2021

Recently, investors' interest and the influence of stock-related information dissemination are being considered as significant factors that explain stock returns and volume. Besides, companies that develop, distribute, or utilize innovative new technologies such as artificial intelligence have a problem that it is difficult to accurately predict a company's future stock returns and volatility due to macro-environment and market uncertainty. Market uncertainty is recognized as an obstacle to the activation and spread of artificial intelligence technology, so research is needed to mitigate this. Hence, the purpose of this study is to propose a machine learning model that predicts the volatility of a company's stock price by using the internet search volume of artificial intelligence-related technology keywords as a measure of the interest of investors. To this end, for predicting the stock market, we using the VAR(Vector Auto Regression) and deep neural network LSTM (Long Short-Term Memory). And the stock price prediction performance using keyword search volume is compared according to the technology's social acceptance stage. In addition, we also conduct the analysis of sub-technology of artificial intelligence technology to examine the change in the search volume of detailed technology keywords according to the technology acceptance stage and the effect of interest in specific technology on the stock market forecast. To this end, in this study, the words artificial intelligence, deep learning, machine learning were selected as keywords. Next, we investigated how many keywords each week appeared in online documents for five years from January 1, 2015, to December 31, 2019. The stock price and transaction volume data of KOSDAQ listed companies were also collected and used for analysis. As a result, we found that the keyword search volume for artificial intelligence technology increased as the social acceptance of artificial intelligence technology increased. In particular, starting from AlphaGo Shock, the keyword search volume for artificial intelligence itself and detailed technologies such as machine learning and deep learning appeared to increase. Also, the keyword search volume for artificial intelligence technology increases as the social acceptance stage progresses. It showed high accuracy, and it was confirmed that the acceptance stages showing the best prediction performance were different for each keyword. As a result of stock price prediction based on keyword search volume for each social acceptance stage of artificial intelligence technologies classified in this study, the awareness stage's prediction accuracy was found to be the highest. The prediction accuracy was different according to the keywords used in the stock price prediction model for each social acceptance stage. Therefore, when constructing a stock price prediction model using technology keywords, it is necessary to consider social acceptance of the technology and sub-technology classification. The results of this study provide the following implications. First, to predict the return on investment for companies based on innovative technology, it is most important to capture the recognition stage in which public interest rapidly increases in social acceptance of the technology. Second, the change in keyword search volume and the accuracy of the prediction model varies according to the social acceptance of technology should be considered in developing a Decision Support System for investment such as the big data-based Robo-advisor recently introduced by the financial sector.

• Information Systems Review
• /
• v.17 no.2
• /
• pp.193-218
• /
• 2015

The purpose of this study is to investigate the effects of closed-type SNS use (i.e., Naver Band) on the perception and behavior of the Korean Army soldiers. In contrast to open-type SNS (e.g., Facebook or Twitter), Naver Band is an online communication service system mostly based on confined offline social network. Therefore, it increases communication between acquaintances who have previously formed relationships. Although the Korean Army recently began to use Naver Band as a method of communication between soldiers, their parents/acquaintance, and Army commanders (or leaders), little research has been done about how this use directly affects army soldiers. Hence, applying the motivation opportunity ability theory of behavior, this study examines how enjoyment (Motivational factor), social ties (Opportunity factor), and social intelligence (Ability factor) affect soldiers' belongingness to their organization and organizational citizenship behavior (OCB). We also hypothesize that army soldiers' belongingness and OCB may enhance their individual performance. Survey results show that enjoyment, social ties, and social intelligence increase army soldiers' belongingness, which leads to OCB. Also, enhanced OCB increases individual performance. However, the effect of enjoyment and social ties on soldiers' OCB is non-significant and soldiers' belongingness does not have influence on individual performance. Theoretical and practical implications are presented.

• Korean Journal of Environment and Ecology
• /
• v.37 no.4
• /
• pp.289-301
• /
• 2023

Forests within watersheds are essential in maintaining ecosystems and are the central infrastructure for constructing an ecological network system. However, due to indiscriminate development projects carried out over past decades, forest fragmentation and land use changes have accelerated, and their original functions have been lost. Since a forest's structural pattern directly impacts ecological processes and functions in understanding forest ecosystems, identifying and analyzing change patterns is essential. Therefore, this study analyzed structural changes in the forest landscape according to the time-series land cover changes using the FRAGSTATS model for the dam watershed of the Geum River upstream. Land cover changes in the dam watershed of the Geum River upstream through land cover change detection showed an increase of 33.12 square kilometers (0.62%) of forests and 67.26 square kilometers (1.26%) of urbanized dry areas and a decrease of 148.25 square kilometers (2.79%) in agricultural areas from the 1980s to the 2010s. The results of no-sampling forest landscape analysis within the watershed indicated landscape percentage (PLAND), area-weighted proximity index (CONTIG_AM), average central area (CORE_MN), and adjacency index (PLADJ) increased, and the number of patches (NP), landscape shape index (LSI), and cohesion index (COHESION) decreased. Identification of structural change patterns through a moving window analysis showed the forest landscape in Sangju City, Gyeongsangbuk Province, Boeun County in Chungcheongbuk Province, and Jinan Province in Jeollabuk Province was relatively well preserved, but fragmentation was ongoing at the border between Okcheon County in Chungcheongbuk Province, Yeongdong and Geumsan Counties in Chungcheongnam Province, and the forest landscape in areas adjacent to Muju and Jangsu Counties in Jeollabuk Province. The results indicate that it is necessary to establish afforestation projects for fragmented areas when preparing a future regional forest management strategy. This study derived areas where fragmentation of forest landscapes is expected and the results may be used as basic data for assessing the health of watershed forests and establishing management plans.

• Spatial Information Research
• /
• v.13 no.3 s.34
• /
• pp.195-209
• /
• 2005

Forest road has been an essential infrastructure for various forestry practices as well as for recreational use, disaster management, and local economics promotion. Since 1980s, extensive network of forest roads has been constructed as an national project in Korea. However, due to the minimal-budget of the project, accurate maps of forest road are not usually available. Although forest road map is a main thematic layer for the forest Geographic Information System (FGIS), its locational accuracy has not been sufficient for the practical applications and, therefore, the update of digital forest road maps is urgent. The objectives of this study is to compare ae methodology of generating and updating digital forest road maps from the aspects of the map accuracy and the efficiency of methods. Four mapping methods (GPS surveying, satellite imagery, ortho aerial photograph, and digital photogrammetry) were applied to generate the forest road maps over the study area of Mt. Oseo in Chungchungnam-do, which has a 35km forest roads distributed in national, public and private forests. The forest road Imp produced by digital photogrammetric method is the most accurate and comparable to GPS surveying although it required the greatest amount of labor time.

• Journal of Digital Convergence
• /
• v.10 no.8
• /
• pp.135-148
• /
• 2012

This work is intended to investigate how the factors of enterprise mobile SNS affect user satisfaction and continuous use intention through technology acceptance model proposed by Davis. To achieve the purpose, this researcher explored Information Systems Success model proposed by DeLone & McLean, Technology Acceptance Model proposed by Davis, and Model after Acceptance, and on the basis of the investigation, performed a study. For the data of this work, 9 enterprises, each of which has more than 100 employees and is located in Seoul, were chosen, and a questionnaire survey was conducted on their 276 employees who experienced enterprise mobile SNS. As a data collection tool, a structured self-administered questionnaire was used. For data analysis, SPSS 18.0 and AMOS 18.0 were used for applying Structural Equation modelling. According to the results of this work, three factors of enterprise mobile SNS-systematic factor (system quality, information quality, and service quality), user factor (personal innovation and personal familiarity), social factor (social effects and social interaction)-affected user satisfaction and continuous use intention through perceived availability, perceived easiness, and perceived enjoyment. Also, it was found that the direction of effects matched a theoretical prediction. And, it was revealed that the decision variables and mediating variables significantly affected user satisfaction and continuous use intention. Theoretical and practical meanings were discussed for the study result, and some suggestions were made for the issues of this work and future studies.

• International Commerce and Information Review
• /
• v.14 no.3
• /
• pp.539-558
• /
• 2012

The purpose of this study is examining the differences of mobile SNS's service quality, which consists data quality and system quality, among the groups that are classified by commitment and customer loyalty. For the experimental analysis, the frequency analysis was performed for general characteristics of sample. The variables were selected by factor analysis that also prove the validity of variables. The value of Cronbach's alpha was calculated to check the reliability of variables. In addition, the group was determined by the both hierarchical and hierarchical cluster analysis, then ANOVA was performed to test the hypotheses that there are differences of mobile SNS's service quality, among the groups that are classified by commitment and customer loyalty. The results of this study support that there are differences among the groups toward mobile SNS's service quality and also shows the more commitment and loyalty group is the higher recognition of mobile SNS's service quality. Thus, the companies have to realize that mobile SNS is very important key factor to success in rapidly changing business environment. In conclusion, the companies implement different customized strategy for the different group and develop the contents and the applications to maximize the commitment and loyalty of for the mobile SNS users.

• International Commerce and Information Review
• /
• v.18 no.3
• /
• pp.3-21
• /
• 2016

The thesis examines genealogical study of various aspects to overcome lots of problems which come by when we execute SWIFTNet TSU BPO. Practical implications regarding the innovation of electronic trade infrastructure are as follows. First, the shipping documents in the SWIFTNet TSU BPO are directly sent to an importer by an exporter after the baseline is confirmed. With this process itself, therefore, the bank cannot secure the account receivable. When initiating the SWIFTNet TSU BPO deal, it is needed to set regulations on the bank's account receivable security in the contract. Second, the SWIFTNet TSU BPO should also have an institutionally unified sharing platform with security, stability and convenience. It other words, it is needed to develop services which meet e-payment paradigm and international environments through continued analysis on market changes and flow. Third, the SWIFTNet TSU is useful in terms of promptness, reduction of risk in foreign exchange payment, cost reduction. Therefore, the SWIFT should be perfectly united and linked among the banks, importer and exporter to make the SWIFTNet TSU more convenient in countries around the world. Fourth, the SWIFT should be approached from the aspect of expansion of network and creation of a new business model through analysis on these problems with a worldwide perspective. At the same time, it is necessary to build a cooperative system to share information and promote comprehensive management for efficient operation.

• Journal of Preventive Medicine and Public Health
• /
• v.31 no.3 s.62
• /
• pp.471-480
• /
• 1998

This study was undertaken in order to estimate the accuracy of disease code of the Korean National Medical Insurance Data and disease the characteristics related to the accuracy. To accomplish these objectives, 2,431 cases coded as notifiable acute communicable diseases (NACD) were randomly selected from 1994 National Medical Insurance data file and family medicine specialists reviewed the medical records to confirm the diagnostic accuracy and investigate the related factors. Major findings obtained from this study are as follows : 1. The accuracy rate of disease code of NACD in National Medical Insurance data was very low, 10.1% (95% C.I. : 8.8-11.4). 2. The reasons of inaccuracy in disease code were 1) claiming process related administrative error by physician and non-physician personnel in medical institutions (41.0%), 2) input error of claims data by key punchers of National Medical Insurer (31.3%) and 3) diagnostic error by physicians (21.7%). 3. Characteristics significantly related with lowering the accuracy of disease code were location and level of the medical institutions in multiple logistic regression analysis. Medical institutions in Seoul showed lower accuracy than those in Kyonngi, and so did general hospitals, hospitals and clinics than tertiary hospitals. Physician related characteristics significantly lowering disease code accuracy of insurance data were sex, age group and specialty. Male physicians showed significantly lower accuracy than female physicians; thirties and fortieg age group also showed significantly lower accuracy than twenties, and so did general physicians and other specialists than internal medicine/pediatric specialists. This study strongly suggests that a series of policies like 1) establishment of peer review organization of National Medical Insurance data, 2) prompt nation-wide expansion of computerized claiming network of National Medical Insurance and 3) establishment and distribution of objective diagnostic criteria to physicians are necessary to set up a national disease surveillance system utilizing National Medical Insurance claims data.

• Information Systems Review
• /
• v.12 no.3
• /
• pp.75-88
• /
• 2010

Business models in open market systems targeting smart phone users are determined by several important factors. First, by providing developers efficient technical platforms, it contains a setting for developers to learn, apply and improve the skills relating to the product category easily while they stay beyond a corporate boundary. Second, by the first condition, a huge population of talented developers becomes to join a specific open market where will invite more customers to use their applications. Hence it will attract more and more developer participants who will finally give a rise to a persistent market growth. Third, the evaluation system between platform providers and application producers, and one between application producers and application users may underlie the trust relationships between them. The research conducted a multiple embedded case study to test the success factors of open market based business models. It focused on smart phone game communities that have installed user evaluation, and feedback systems. The user innovation empowerment model within the social game networks has highlighted the theories on the roles and characteristics of lead users, and lead user network behaviors for future NPD participations.