• Journal of the Korean Operations Research and Management Science Society
• /
• v.34 no.2
• /
• pp.35-54
• /
• 2009

The recent economic crisis not only reduces the profit of department stores but also incurs the significance losses caused by the increasing late-payment rate of credit cards. Under this pressure, the scope of credit prediction needs to be broadened from the simple prediction of whether this customer has a good credit or not to the accurate prediction of how much profit can be gained from this customer. This study classifies the delinquent customers of credit card in a Korean department store into homogeneous clusters. Using this information, this study analyzes the repayment patterns for each cluster and develops the credit prediction system to manage the delinquent customers. The model presented by this study uses Kohonen network, which is one of artificial neural networks of data mining technique, to cluster the credit delinquent customers into clusters. Cox proportional hazard model is also used, which is one of survival analysis used in medical statistics, to analyze the repayment patterns of the delinquent customers in each cluster. The presented model estimates the repayment period of delinquent customers for each cluster and introduces the influencing variables on the repayment pattern prediction. Although there are some differences among clusters, the variables about the purchasing frequency in a month and the average number of installment repayment are the most predictive variables for the repayment pattern. The accuracy of the presented system leaches 97.5%.

• Genomics & Informatics
• /
• v.14 no.4
• /
• pp.138-148
• /
• 2016

The success of genome-wide association studies (GWASs) has enabled us to improve risk assessment and provide novel genetic variants for diagnosis, prevention, and treatment. However, most variants discovered by GWASs have been reported to have very small effect sizes on complex human diseases, which has been a big hurdle in building risk prediction models. Recently, many statistical approaches based on penalized regression have been developed to solve the "large p and small n" problem. In this report, we evaluated the performance of several statistical methods for predicting a binary trait: stepwise logistic regression (SLR), least absolute shrinkage and selection operator (LASSO), and Elastic-Net (EN). We first built a prediction model by combining variable selection and prediction methods for type 2 diabetes using Affymetrix Genome-Wide Human SNP Array 5.0 from the Korean Association Resource project. We assessed the risk prediction performance using area under the receiver operating characteristic curve (AUC) for the internal and external validation datasets. In the internal validation, SLR-LASSO and SLR-EN tended to yield more accurate predictions than other combinations. During the external validation, the SLR-SLR and SLR-EN combinations achieved the highest AUC of 0.726. We propose these combinations as a potentially powerful risk prediction model for type 2 diabetes.

• 한국경영정보학회:학술대회논문집
• /
• 2007.06a
• /
• pp.608-613
• /
• 2007

Prediction of corporate bankruptcies has long been an important topic and has been studied extensively in the finance and management literature because it is an essential basis for the risk management of financial institutions. Recently, support vector machines (SVMs) are becoming popular as a tool for bankruptcy prediction because they use a risk function consisting of the empirical error and a regularized term which is derived from the structural risk minimization principle. In addition, they don't require huge training samples and have little possibility of overfitting. However. in order to Use SVM, a user should determine several factors such as the parameters ofa kernel function, appropriate feature subset, and proper instance subset by heuristics, which hinders accurate prediction results when using SVM In this study, we propose a novel hybrid SVM classifier with simultaneous optimization of feature subsets, instance subsets, and kernel parameters. This study introduces genetic algorithms (GAs) to optimize the feature selection, instance selection, and kernel parameters simultaneously. Our study applies the proposed model to the real-world case for bankruptcy prediction. Experimental results show that the prediction accuracy of conventional SVM may be improved significantly by using our model.

• Journal of Positioning, Navigation, and Timing
• /
• v.9 no.2
• /
• pp.89-98
• /
• 2020

Prediction of clock behaviors is necessary to generate very high stable system time which is essential for a satellite navigation system. For the purpose, we applied the Auto-Regressive Integrated Moving Average (ARIMA) model to the prediction of two hydrogen masers' behaviors with respect to the rapid Coordinated Universal Time (UTCr). Using the packaged programming language R, we made an analysis and prediction of time series data of [UTCr - clocks]. The maximum variation width of the residuals which were obtained by the difference between the predicted and measured values, was 6.2 ns for 106 days. This variation width was just one-sixth of [UTCr-UTC (KRIS)] published by the BIPM for the same period. Since the two hydrogen masers were found to be strongly correlated, we applied the Vector Auto-Regressive Moving Average (VARMA) model for more accurate prediction. The result showed that the prediction accuarcy was improved by two times for one hydrogen maser.

• Journal of the Korea Institute of Building Construction
• /
• v.2 no.4
• /
• pp.177-182
• /
• 2002

The purpose of this study is to develope the strength prediction model by Maturity Method. A maturity function is a mathematical expression to account for the combined effects of time and temperature on the strength development of a cementious mixture. The method of equivalent ages is to use Arrhenius equation which indicates the influence of curing temperature on the initial hydration ratio of cement. For the experimental factors of this study, we selected the concrete mixing of W/C ratio 45, 50, 55 and 60% and curing temperature 5, 10, 20 and $30^{\circ}C$. And we compare and evaluate with logistic model that is existing strength prediction model, because we have to verify adaption possibility of new strength prediction model which is proposed by maturity method. As the results, it is found that investigation of the activation energy that are used to calculate equivalent age is necessary, and new strength prediction model was proved to be more accurate in the strength prediction than logistic model in the early age. Moreover, the use of new model was more reasonable because it has low SSE and high decisive factor.

• Journal of the Korean GEO-environmental Society
• /
• v.20 no.10
• /
• pp.29-38
• /
• 2019

In stability and settlement management of soft ground, the settlement prediction technology has been continuously developed and used to reduce construction cost and confirm the exact land use time. However, the preexistence prediction methods such as hyperbolic method, Asaoka method and Hoshino method are difficult to predict the settlement accurately at the beginning of consolidation because the accurate settlement prediction is possible only after many measurement periods have passed. It is judged as the reason for estimating the future settlement through the proportionality assumption of the slope which the preexistence prediction method computes from the settlement curve. In this study, ARIMA technique is introduced among time series analysis techniques and compared with preexistence prediction methods. ARIMA method was predictable without any distinction of ground conditions, and the results similar to the existing method are predicted early (final settlement).

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.12
• /
• pp.29-38
• /
• 2019

A path prediction method using lifelog requires a large amount of training data for accurate path prediction, and the path prediction performance is degraded when the training data is insufficient. The lack of training data can be solved using data of other users having similar user movement patterns. Therefore, this paper proposes a path prediction algorithm based on user similarity. The proposed algorithm learns the path in a triple grid pattern and measures the similarity between users using the cosine similarity technique. Then, it predicts the path with applying measured similarity to the learned model. For the evaluation, we measure and compare the path prediction accuracy of proposed method with the existing algorithms. As a result, the proposed method has 66.6% accuracy, and it is evaluated that its accuracy is 1.8% higher than other methods.

• International Journal of Computer Science & Network Security
• /
• v.22 no.10
• /
• pp.171-176
• /
• 2022

A dramatic rise in the number of people dying from heart disease has prompted efforts to find a way to identify it sooner using efficient approaches. A variety of variables contribute to the condition and even hereditary factors. The current estimate approaches use an automated diagnostic system that fails to attain a high level of accuracy because it includes irrelevant dataset information. This paper presents an effective neural network with convolutional layers for classifying clinical data that is highly class-imbalanced. Traditional approaches rely on massive amounts of data rather than precise predictions. Data must be picked carefully in order to achieve an earlier prediction process. It's a setback for analysis if the data obtained is just partially complete. However, feature extraction is a major challenge in classification and prediction since increased data increases the training time of traditional machine learning classifiers. The work integrates the CNN-MDRP classifier (convolutional neural network (CNN)-based efficient multimodal disease risk prediction with TANFIS (tuned adaptive neuro-fuzzy inference system) for earlier accurate prediction. Perform data cleaning by transforming partial data to informative data from the dataset in this project. The recommended TANFIS tuning parameters are then improved using a Laplace Gaussian mutation-based grasshopper and moth flame optimization approach (LGM²G). The proposed approach yields a prediction accuracy of 98.40 percent when compared to current algorithms.

• Geomechanics and Engineering
• /
• v.30 no.5
• /
• pp.413-422
• /
• 2022

TBM is widely used in the construction of various underground projects in the current world, and has the unique advantages that cannot be compared with traditional excavation methods. However, due to the high cost of TBM, the damage is even greater when geological disasters such as collapse occur during excavation. At present, there is still a shortage of research on various types of risk prediction of TBM tunnel, and accurate and reliable risk prediction model is an important theoretical basis for timely risk avoidance during construction. In this paper, a prediction model is proposed to evaluate the risk level of tunnel collapse by establishing a reasonable risk index system, using analytic hierarchy process to determine the index weight, and using the normal cloud model theory. At the same time, the traditional analytic hierarchy process is improved and optimized to ensure the objectivity of the weight values of the indicators in the prediction process, and the qualitative indicators are quantified so that they can directly participate in the process of risk prediction calculation. Through the practical engineering application, the feasibility and accuracy of the method are verified, and further optimization can be analyzed and discussed.

• Architectural research
• /
• v.24 no.1
• /
• pp.1-11
• /
• 2022

As the occurrence of a crime is dependent on different factors, their correlations are beyond the ordinary cognitive range. Owing to this limitation, systems face difficulty in correlating various factors, thereby requiring the assistance of artificial intelligence (AI) to overcome such limitations. Therefore, AI has become indispensable for crime prediction. Crimes can cause severe and irrevocable damage to a society. Recently, big data has been introduced for developing highly accurate models for crime prediction. Prediction of night crimes should be given significant consideration, because crimes primarily occur during nights, when the spatiotemporal characteristics become vulnerable to crimes. Many environmental factors that influence crime rate are applied for crime prediction, and their influence on crime rate may differ based on temporal characteristics and the nature of crime. This study aims to identify the environmental factors that influence sex and theft crimes occurring at night and proposes an artificial neural network (ANN) model to predict sex and theft crimes at night in random areas. The crime data of A district in Seoul for 12 years (2004-2015) was used, and environmental factors that influence sex and theft crimes were derived through multiple regression analysis. Two types of crime prediction models were developed: Type A using all environmental factors as input data; Type B with only the significant factors (obtained from regression analysis) as input data. The Type B model exhibited a greater accuracy than Type A, by 3.26 and 9.47 % higher for theft and sex crimes, respectively.