• Journal of Digital Convergence
• /
• v.20 no.4
• /
• pp.185-192
• /
• 2022

This study investigates the effect of the semi-supervised learning(SSL) method on predicting default risk of peer-to-peer(P2P) loans. Despite its proven performance, the supervised learning(SL) method requires labeled data, which may require a lot of effort and resources to collect. With the rapid growth of P2P platforms, the number of loans issued annually that have no clear final resolution is continuously increasing leading to abundance in unlabeled data. The research data of P2P loans used in this study were collected on the LendingClub platform. This is why an SSL model is needed to predict the default risk by using not only information from labeled loans(fully paid or defaulted) but also information from unlabeled loans. The results showed that in terms of default risk prediction and despite the use of a small number of labeled data, the SSL method achieved a much better default risk prediction performance than the SL method trained using a much larger set of labeled data.

• 한국방재학회:학술대회논문집
• /
• 2008.02a
• /
• pp.645-648
• /
• 2008

This study provides considerations for quantitative risk assessment of landslide on GIS technology. It shows how the landslide possibility analysis is linked by GIS modeling to provide loss estimation tools for landslide hazards in support of socio-economic loss reduction efforts. Those risk assessment results can deliver factual damage situation prediction to policy making for the landslide damage mitigation.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.20 no.6
• /
• pp.56-64
• /
• 2016

Damage of sea-crossing bridge by ship collision is related to estimate frequencies of overloading due to impact, and bridge accordingly must be designed to satisfy related acceptance criteria. Another important aspect is the management on increment of collision risk during the service period. In this study, related plan, main span length, air draft clearance and collision risk are analyzed for the interim assessment of Incheon Bridge focusing on the ship collision problem. In particular, for the increment of collision risk, the optimized navigation speed is proposed by reviewing the research findings and navigation guidelines etc. as a temporary expedient. Also basic procedure for reasonable prediction of target vessel and passage is established and probabilistic prediction method to embrace the uncertainty of the prediction is proposed as a fundamental solution. It is necessary to conduct further research on collision risk management and promptly carry out interim assessments of other marine bridges.

• Fire Science and Engineering
• /
• v.14 no.4
• /
• pp.7-16
• /
• 2000

In the study, three criteria(toxicity, fire & explosion, environment) and damage prediction method for each case was set up, and all these criteria were applied to the subject substance that was selected as hazardous level by integrating all criteria through Algorithm. Particularly, the environment criterion is a comprehensive concept, environment index modeling by combining USCG(United State Coast Guard) & MSDS(Material Safety Data Sheet) environment criteria classifications and the environment part of MFPA's health hazardousnes(Nh). And for damage prediction method of each criterion were adopted and they were applied to hazardous chemical substances in use or stored by chemical substance related enterprises located in each region that made possible to set up total hazard level of used substances(inflammability, poisonousness and counteraction on a unit substance, and hazard level & display modeling on environment) & damage prediction in case of accident & solidity setup(CPQRA: Chemical Process Quantitative Risk Assessment, IAEA: International Atomic Energy Agency, VZ eq: Vulnerable Zone) risk counter. Thus it is deemed that it can be applied to toxic substance leakage that can happen during any chemical processing & storage, application as a tool for prior safety evaluation through potential dangerousness computation of fire & explosion.

• Convergence Security Journal
• /
• v.14 no.6_2
• /
• pp.89-95
• /
• 2014

This study analyzed a correlation between crime and meteorological changes and discomfort index of Seoul and p resented a prediction expression through the regression analysis. For data used in this study, crime data from Januar y 2008 to December 2012 of Seoul Metropolitan Police Agency and meteorological records and discomfort index recor ded in the Meteorological Agency through the portal sites were used. Based on this data, SPSS 18.0 was used for the regression analysis and the analysis of correlation between crime and meteorological changes and discomfort index and a prediction expression was derived through the analysis and the risk index was shown in 5 steps depending on predicted values obtained through the prediction expression derived. The risk index of 5 steps classified like this is considered to be used as important data for crime prevention activities.

• Korean Management Science Review
• /
• v.33 no.3
• /
• pp.105-117
• /
• 2016

In this paper, we present empirical testing result to examine the validity of inbound supply and outbound demand risk factors in the sense of early predicting the firm's bankruptcy risk level. The risk factors are drawn from industry uncertainty attributes categorized as uncertainties of input market (inbound supply), and product market (outbound demand). On the basis of input-output table, industry level inbound and outbound sectors are identified to formalize supply chain structures, relevant inbound and outbound uncertainty attributes and corresponding risk factors. Subsequently, publicly available macro-economic indicators are used to appropriately quantify these risk factors. Total 68 industry level bankruptcy risk forecasting results are presented with the average R-square scores of between 53.4% and 37.1% with varying time lag. The findings offers useful insights to incorporate supply chain risk to the body of firm's bankruptcy risk level prediction literature.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.1
• /
• pp.595-598
• /
• 2013

Background: Lung cancer is the leading cause of cancer death worldwide Therefore, identification of genetic as well as environmental factors is very important in developing novel methods of lung cancer prevention. However, this is a multi-layered problem. Therefore a lung cancer risk prediction system is here proposed which is easy, cost effective and time saving. Materials and Methods: Initially 400 cancer and non-cancer patients' data were collected from different diagnostic centres, pre-processed and clustered using a K-means clustering algorithm for identifying relevant and non-relevant data. Next significant frequent patterns are discovered using AprioriTid and a decision tree algorithm. Results: Finally using the significant pattern prediction tools for a lung cancer prediction system were developed. This lung cancer risk prediction system should prove helpful in detection of a person's predisposition for lung cancer. Conclusions: Most of people of Bangladesh do not even know they have lung cancer and the majority of cases are diagnosed at late stages when cure is impossible. Therefore early prediction of lung cancer should play a pivotal role in the diagnosis process and for an effective preventive strategy.

• Genomics & Informatics
• /
• v.11 no.1
• /
• pp.46-51
• /
• 2013

Normal-karyotype acute myeloid leukemia (NK-AML) is a highly malignant and cytogenetically heterogeneous hematologic cancer. We searched for somatic mutations from 10 pairs of tumor and normal cells by using a highly efficient and reliable analysis workflow for whole-exome sequencing data and performed association tests between the NK-AML and somatic mutations. We identified 21 nonsynonymous single nucleotide variants (SNVs) located in a coding region of 18 genes. Among them, the SNVs of three leukemia-related genes (MUC4, CNTNAP2, and GNAS) reported in previous studies were replicated in this study. We conducted stepwise genetic risk score (GRS) models composed of the NK-AML susceptible variants and evaluated the prediction accuracy of each GRS model by computing the area under the receiver operating characteristic curve (AUC). The GRS model that was composed of five SNVs (rs75156964, rs56213454, rs6604516, rs10888338, and rs2443878) showed 100% prediction accuracy, and the combined effect of the three reported genes was validated in the current study (AUC, 0.98; 95% confidence interval, 0.92 to 1.00). Further study with large sample sizes is warranted to validate the combined effect of these somatic point mutations, and the discovery of novel markers may provide an opportunity to develop novel diagnostic and therapeutic targets for NK-AML.

• Journal of Navigation and Port Research
• /
• v.28 no.1
• /
• pp.17-22
• /
• 2004

The paper describes implementation of three-dimensional visualization system that is to provide comprehensive meaning of the statistical prediction results on the marine casualties. Graphical User Interface (GUI) and Web based Virtual Reality (VR) technology are mainly introduced in the system development. To provide daily forecasting, time based casualty prediction model and risk level index are developed in this work. As operating test results of the system, complicated statistical meaning can be shown in the three-dimensional virtual space using simple color. In addition, daily risk levels can be shown on the bar-graph.

• IE interfaces
• /
• v.20 no.4
• /
• pp.448-457
• /
• 2007

The small & micro business has the characteristics of both consumer credit risk and business credit risk. In predicting the bankruptcy for small-micro businesses, the problem is that in most cases, the financial data for evaluating business credit risks of small & micro businesses are not available. To alleviate such problem, we propose a bankruptcy prediction mechanism using the credit card sales information available, because most small businesses are member store of some credit card issuers, which is the main purpose of this study. In order to perform this study, we derive some variables and analyze the relationship between good and bad signs. We employ the new statistical learning technique, support vector machines (SVM) as a classifier. We use grid search technique to find out better parameter for SVM. The experimental result shows that credit card sales information could be a good substitute for the financial data for evaluating business credit risk in predicting the bankruptcy for small-micro businesses. In addition, we also find out that SVM performs best, when compared with other classifiers such as neural networks, CART, C5.0 multivariate discriminant analysis (MDA), and logistic regression.