• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.25-38
• /
• 2019

Selecting high-quality information that meets the interests and needs of users among the overflowing contents is becoming more important as the generation continues. In the flood of information, efforts to reflect the intention of the user in the search result better are being tried, rather than recognizing the information request as a simple string. Also, large IT companies such as Google and Microsoft focus on developing knowledge-based technologies including search engines which provide users with satisfaction and convenience. Especially, the finance is one of the fields expected to have the usefulness and potential of text data analysis because it's constantly generating new information, and the earlier the information is, the more valuable it is. Automatic knowledge extraction can be effective in areas where information flow is vast, such as financial sector, and new information continues to emerge. However, there are several practical difficulties faced by automatic knowledge extraction. First, there are difficulties in making corpus from different fields with same algorithm, and it is difficult to extract good quality triple. Second, it becomes more difficult to produce labeled text data by people if the extent and scope of knowledge increases and patterns are constantly updated. Third, performance evaluation is difficult due to the characteristics of unsupervised learning. Finally, problem definition for automatic knowledge extraction is not easy because of ambiguous conceptual characteristics of knowledge. So, in order to overcome limits described above and improve the semantic performance of stock-related information searching, this study attempts to extract the knowledge entity by using neural tensor network and evaluate the performance of them. Different from other references, the purpose of this study is to extract knowledge entity which is related to individual stock items. Various but relatively simple data processing methods are applied in the presented model to solve the problems of previous researches and to enhance the effectiveness of the model. From these processes, this study has the following three significances. First, A practical and simple automatic knowledge extraction method that can be applied. Second, the possibility of performance evaluation is presented through simple problem definition. Finally, the expressiveness of the knowledge increased by generating input data on a sentence basis without complex morphological analysis. The results of the empirical analysis and objective performance evaluation method are also presented. The empirical study to confirm the usefulness of the presented model, experts' reports about individual 30 stocks which are top 30 items based on frequency of publication from May 30, 2017 to May 21, 2018 are used. the total number of reports are 5,600, and 3,074 reports, which accounts about 55% of the total, is designated as a training set, and other 45% of reports are designated as a testing set. Before constructing the model, all reports of a training set are classified by stocks, and their entities are extracted using named entity recognition tool which is the KKMA. for each stocks, top 100 entities based on appearance frequency are selected, and become vectorized using one-hot encoding. After that, by using neural tensor network, the same number of score functions as stocks are trained. Thus, if a new entity from a testing set appears, we can try to calculate the score by putting it into every single score function, and the stock of the function with the highest score is predicted as the related item with the entity. To evaluate presented models, we confirm prediction power and determining whether the score functions are well constructed by calculating hit ratio for all reports of testing set. As a result of the empirical study, the presented model shows 69.3% hit accuracy for testing set which consists of 2,526 reports. this hit ratio is meaningfully high despite of some constraints for conducting research. Looking at the prediction performance of the model for each stocks, only 3 stocks, which are LG ELECTRONICS, KiaMtr, and Mando, show extremely low performance than average. this result maybe due to the interference effect with other similar items and generation of new knowledge. In this paper, we propose a methodology to find out key entities or their combinations which are necessary to search related information in accordance with the user's investment intention. Graph data is generated by using only the named entity recognition tool and applied to the neural tensor network without learning corpus or word vectors for the field. From the empirical test, we confirm the effectiveness of the presented model as described above. However, there also exist some limits and things to complement. Representatively, the phenomenon that the model performance is especially bad for only some stocks shows the need for further researches. Finally, through the empirical study, we confirmed that the learning method presented in this study can be used for the purpose of matching the new text information semantically with the related stocks.

• 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 Science Education
• /
• v.39 no.3
• /
• pp.358-375
• /
• 2015

The purposes of this study were to investigate the value, necessity and legitimacy of STEAM Education and to propose practical approaching methods for STEAM Education to be applicable in Korea through a variety of literature review, case studies and collecting suggestions from domestic and international educational experts. The research questions are as follows: (1) To investigate the perception, understanding and recognitions of domestic and foreign professionals in STEAM education. (2) To analyze policy implications for an improvement in STEAM. The following aspects of STEAM were found to be challenges in our current STEAM policy after analyzing multiple questionnaires with the professionals and case studies including their experiences, understanding, supports and directions of the policy from the governments. The results indicate that (1) there was a lack of precise and conceptual understanding of STEAM in respect to experience. Training sessions for teachers in this field to help transform their perception is necessary. Development of practical programs with an easy access is also required. It is important to get the aims of related educational activities recognized by the professionals and established standards for an evaluation. The experts perceived that a theme-based learning is the most preferred and effective approaching method and the programs that develop creative thinking and learning applicable to practice are required to promote. (2) The results indicate that there was a lack of programs and inducements for supporting outstanding STEAM educators. It is shown that making an appropriate environment for STEAM education takes the first priority before training numbers of teachers unilaterally, thus securing enough budget seems critical. The professionals also emphasize on developing specialized teaching materials that include diverse inter-related subjects such as science technology, engineering, arts and humanities and social science with diverse viewpoints and advanced technology. This work requires a STEAM network for teachers to link up and share their materials, documents and experiences. It is necessary to get corporations, universities, and research centers participated in the network. (3) With respect to direction, it is necessary to propose policy that makes STEAM education ordinary and more practical in the present education system. The professionals have recommended training sessions that help develop creative thinking and amalgamative problem-solving techniques. They require reducing the workload of teachers and changing teachers' perspectives towards STEAM. They further urge a tight cooperation between departments of the government related with STEAM.

• Archives of design research
• /
• v.18 no.4 s.62
• /
• pp.95-104
• /
• 2005

The purpose of this study is to estimate the determinants of product design and analyze the impacts of product design on quality competitiveness, product reliability, and consumer satisfaction in an attempt to provide a foundation for the theory of design management. For this empirical analysis, this study has derived the relevant measurement variables from a survey on 400 Korean manufacturing firms during the period of $August{\sim}October$ 2003. The empirical findings are summarized as follows: First, the determinants of product design are very significantly (at p<0.001) estimated to be the R&D capability, the level of R&D expenditure, the level of innovative activities(5S, TQM, 6Sigma, QC, etc.). This empirical result can support Pawar and Driva(1999)'s two principles by which the performance of product design and product development can be simultaneously evaluated in the context of CE(concurrent engineering) of NPD(newly product development) activities. Second, the hypothesis on the causality: product design${\rightarrow}$quality competitiveness${\rightarrow}$customer satisfaction${\rightarrow}$customer loyalty is very significantly (at p<0.001) accepted. This implies that product design positively affects consumer satisfaction, not directly but indirectly, by influencing quality competitiveness. This empirical result of this study can also support the studies of for example Flynn et al.(1994), Ahire et at.(1996), Afire and Dreyfus(2000) which conclude that design management is a significant determinant of product quality. The aforementioned empirical results are important in the following sense: the empirical result that quality competitiveness plays a bridging role between product design and consumer satisfaction can reconcile the traditional debate between QFD(quality function development) approach asserted by product developers and conjoint analysis maintained by marketers. The first empirical result is related to QFD approach whereas the second empirical result is related to conjoint analysis. At the same time, the empirical results of this study can support the rationale of design integration(DI) of Ettlie(1997), i.e., the coordination of the timing and substance of product development activities performed by the various disciplines and organizational functions of a product's life cycle. Finally, the policy implication (at the corporate level) from the empirical results is that successful design management(DM) requires not only the support of top management but also the removal of communication barriers, (i.e. the adoption of cross-functional teams) so that concurrent engineering(CE), the simultaneous development of product and process designs can assure product development speed, design quality, and market success.

• The korean journal of orthodontics
• /
• v.31 no.5 s.88
• /
• pp.517-523
• /
• 2001

Titanium miniscrews we being used increasingly as an anchorage for tooth movement, because they ate easy to place and to remove, increase the number of sites available, give minimum strain to patients regarding surgical procedures, and offer uneventful healing alter removal. The use of titanium miniscrews as an orthodontic anchorage has been reported in clinical case reports, but clinicians have experienced screw loosening when using such screws.' To our knowledge, there are no published reports evaluating the stability of miniscrews. Information about the length of miniscrews used in relation to the location is of some importance, as stability will vary depending on bone duality The purpose of this study was to evaluate a variety of Lengths of miniscrews (dimeter: 2mm) which were inserted in maxilla or mandible and to demonstrate in a dog model which miniscrew provides fundamental stability in the jaws. 10 mm long miniscrews in the maxilla and 8mm long: miniscrews in the mandible showed no clinical mobility and retained their position throughout an 8 weeks force (200g) application. The mucosal condition around the screws was healthy in cases in which miniserews were inserted in the alveolar bone between the roots and the head of the screws emerged into the attached gingiva. When the force application was terminated, radiographic analysis revealed neither rent resorption not periodontal pathology around the miniscrews that remained stable during the entire treatment period. This study suggests that if titanium miniscrews with adequate length are properly used depending on the location, they provide sufficient stability for orthodontic anchorage.

• Journal of Venture Innovation
• /
• v.1 no.2
• /
• pp.47-59
• /
• 2018

With the recent, rapid changes in the socio-economic environment, organizations of today are now required to present a framework of realistic consumer behaviors based on psychology, economy, and finance, in order to understand their investing customers. Stock investors show differences in terms of their decisions or evaluations in the process of investing. This is due to what is called the 'framing effect.' The decision frames of the investors are defined differently, and, as a result, this affects the decisions made by the investors. Preceding studies on stock investment rarely touched the topic of the effect of message framing on market participants in their stock investment, especially regarding the differences in terms of their risk management behaviors based on the message framing in stock investment. Therefore, the purpose of this study is to examine the influence of stock investment message framing on market participants in their investment decision making and empirically validate whether this message framing effect has a moderating effect on the factors of investment satisfaction. For this, 494 participants with stock investment experiences were interviewed from May 1 to 26, 2018, and the results were used as the data for the empirical analysis. The analysis of the data was conducted using SPSS 22.0 statistical analysis software. The results of this study were as follows; First, of the stock investment behavioral factors, the stock comprehension, recommendation by others for a stock, and the degree of risks of a stock affected stock investment satisfaction in a positive manner. And, of the behavioral factors of stock investment, stock comprehension, stock brand, recommendation on the stocks from others, past performances, and risk levels of stocks affected the intent of continued stock investment in a positive manner. Second, message framing turned out to affect stock investment satisfaction in a positive manner, and it also had a significant moderating effect to the relationship between the stock investment behavior and stock investment satisfaction. Third, message framing was found to affect continued stock investment intent significantly, with a significant moderating effect in the relationship between stock investment behavioral factor and continued stock investment intent.

• Korean Security Journal
• /
• no.10
• /
• pp.53-77
• /
• 2005

This thesis analyzes the effectiveness of master's transformational and transactional leadership on organizational commitment as well as job satisfaction of private security, scrutinizes the difference in master's leadership according to social-demographic group. The sum of this thesis is as follows. To begin with, analyzed the difference of leadership style according to social-demographic characteristics of private security, it showed that monthly income is over 2.5 million won as well as people over 30 years old have higher transformational leadership than the others. That is, the older people's age is and the more someone gains his living, the higher he esteems and trusts his master who has charisma, personal concern and shows intellectual stimulus. And examined thoroughly the difference of organizational commitment as well as job satisfaction in accordance with social-demographic characteristics of private security, people who are 20-25 years old, are college graduates and employees who have worked for 2-3 years do well calculational commitment. It means that the younger he is and the lower academic background is, the higher calculational commitment is by the profit and loss which affects them. Secondly, inquired into consequence of leadership style on job satisfaction. Leadership of intellectual stimulus has negative effect on job satisfaction, whereas it has positive effect on conditional compensation. Exactly, if the master exhibits leadership that stimulate to improve an initiative, it reduces the job satisfaction. On the contrary, for the reason that his master indicates leadership that accompanies conditional compensation, job satisfaction tends to increase. Finally, judged from the effects of leadership style on organizational commitment, the charisma, the lower factor of transformational leadership affects to emotional commitment positively and then, it shows that employees tend to become attached to the organization and have a sense of oneness with it, if their master is able to show charismatic leadership. In conclusion, the private security companies pursue employees whose school career and payment is lower because of incidence in labor expense related with the profit and loss. Owing to it, employees are not satisfied with their job much. And also there is the change in increasing rate of job satisfaction according to business performance of the conditional compensation, i.e. a consideration, a promotion, etc., and the calculational commitment. Therefore, the university should try to focus on the researches for cultivate more professional manpower for business, the companies must try to make better deal, offer welfare for their employees to develop of the private security industry, to rise the job satisfaction of private security. And then, investigation to find private security's own level ought to be done by organic industry-academic cooperation of university and industries.

• Journal of Food Hygiene and Safety
• /
• v.20 no.4
• /
• pp.253-257
• /
• 2005

This study was performed to develop management system of artificial flavor in Korea that considered the usage and management of artificial flavor within or outside (Europe, USA and JECFA) and to offer a yardstick for judgement and prevent from confusing when manufacture or import artificial flavoring substances. In questionnaire survey for flavoring manufacture form, ideal management system and others in companies related artificial flavor, the replier answered that artificial flavor was mainly used to drinks as water soluble from and that the countries exporting flavoring substances most frequently to Korea were Japan. Europe and America sequentially. On the basis of above results, we prepared the positive list (proposal) on about 1800 artificial flavoring substances for application to regulations in Korea Food Additives Code.

• Journal of Technology Innovation
• /
• v.24 no.4
• /
• pp.55-99
• /
• 2016

The paper investigates X-project, in which the public was invited to participate in a national R&D project, examines how X-project attracted the public's attention and involved them in a national R&D project, and discusses the significance and limitations of X-project. X-project was executed by a 12 citizen-led committee, financially supported by the Ministry of Science, ICT, and Future Planning, and backed by the Science and Technology Policy Institute. People raised 6,212 questions that reflected the severe needs they experienced in their daily lives through the online and offline platforms of X-project. In addition, the committee members, scholars, experts, government officials, and citizens gathered together to select the fifty most provocative and novel of the questions raised by the public, and invited public participation to answer the questions in innovative ways. 310 research teams including professional researchers from universities and institutes, high-school students, lay persons, and corporate workers applied for X-project, and 54 of these teams were finally selected to receive funding from the government. Through planning and conducting X-project, as well as interviewing and surveying the participants in X-project and non-participants, we found that there was an enormous social consensus on the necessity of public participatory national R&D projects. People asserted that science and technology should put a greater focus on solving social problems and satisfying public needs. We also confirmed that the public could take part in national R&D projects. Most of all, we found that the questions raised by the public were very challenging, novel, and complex, and thus researchers need break-through approaches to address them. It can be also argued that through experiencing the X-project citizens can regard themselves as ones who are not only recipients of the benefits of the development of science and technology, but also contributors of the development of them. We finally argue that there are some limitations to X-project in terms of how to provide diverse incentives that attract more participation, how to develop the process in which people got involved in the project in more easy ways, and how to create new ways for lay persons and professional researchers to cooperate in solving social problems.

• Korean Security Journal
• /
• no.22
• /
• pp.91-111
• /
• 2010

Around the middle of the ninth century the strict bone-rank system of Silla frustrated many people who had political ambition but lacked nobility. They had to seek other ways, including maritime trade. Such an undertaking reflected and also increased their economic and military power. Trade prospered with T'ang China and with Japan as well. The threat of piracy to Silla's thriving maritime trade caused to create a succession of garrisons at important coastal points. Chonghae Jin (Chonghae garrison) was regarded as the most important of these. It was established in 828 by Chang Pogo. Chonghae Jin was on Wando, an island just east of the southwestern tip of Korea and a key place at this time in the trade between China, Korea, and Japan. From this vantage point Chang Pogo became a merchant-prince with extensive holdings and commercial interests in China and with trade contacts with Japan. Although piracy was rampant in East Asia at that time, either the Chinese or Silla government was not able to control it due to inner political strife and lack of policing resources. Infuriated by the piracy and the government's inability to control it, Chang Pogo came back to Silla to fight against the pirates and to protect maritime trade. He persuaded the king of Silla and was permitted to control the private armed forces to sweep away the pirates. In 829 he was appointed Commissioner of Chonghae-Jin with the mission of curbing piracy in that region. Chang's forces were created to protect people from pirates, but also developed into traders among Silla Korea, T'ang China, and Japan in the 9th century. This was geographically possible because the Chonghae Garrison was situated at the midpoint of Korea, China, and Japan, and also because Chang's naval forces actually dominated the East Asia Sea while patrolling sea-lanes. Based on these advantages, Chang Pogo made a great fortune, which might be collected from a charge for protecting people from pirates and the trades with China and Japan. Chang's forces could be termed the first private security company in the Korean history, at least in terms of historical documents. Based on historical documents, the numbers of private soldiers might be estimated to exceed tens of thousands at least, since Chang's forces alone were recorded to be more than ten thousand. Because local powers and aristocratic elites were said to have thousands of armed forces respectively, the extent of private forces was assumed to be vast, although they were available only to the privileged class. In short, the domination of Chang's forces was attributable to the decline of central government and its losing control over local powers. In addition it was not possible without advanced technologies in shipbuilding and navigation.