• Journal of the Korean Operations Research and Management Science Society
• /
• v.23 no.3
• /
• pp.185-207
• /
• 1998

This research investigates computer generated hybrid second-order model of two numerically based approaches to risk classification : discriminant analysis and neural networks. The hybrid second-order models are derived by rule induction using the ID3 and tested in the several different kinds of data. This new hybrid approach is designed to combine the high prediction accuracy and robustness of DA or NN with perspicuity of ID3. The hybrid model also eliminates the problem of contradictory inputs of ID3. After doing empirical test for the validity of hybrid model using small and medium companies' bankrupt data, hybrid model shows high perspicuity, high prediction accuracy for bankrupt, and simplicity for rules. The hybrid model also shows high performance regardless the type of data such as numeric data, non-numeric data, and combined data.

• Journal of Intelligence and Information Systems
• /
• v.19 no.1
• /
• pp.1-17
• /
• 2013

In this article, several types of hybrid forecasting models are suggested. In particular, hybrid models using the generalized additive model (GAM) are newly suggested as an alternative to those using neural networks (NN). The prediction performances of various hybrid and non-hybrid models are evaluated using simulated time series data. Five different types of seasonal time series data related to an additive or multiplicative trend are generated over different levels of noise, and applied to the forecasting evaluation. For the simulated data with only seasonality, the autoregressive (AR) model and the hybrid AR-AR model performed equivalently very well. On the other hand, if the time series data employed a trend, the SARIMA model and some hybrid SARIMA models equivalently outperformed the others. In the comparison of GAMs and NNs, regarding the seasonal additive trend data, the SARIMA-GAM evenly performed well across the full range of noise variation, whereas the SARIMA-NN showed good performance only when the noise level was trivial.

• Smart Media Journal
• /
• v.5 no.4
• /
• pp.75-82
• /
• 2016

Although Korea's economic and domestic automobile market through the change of road environment are growth, the traffic accident rate has also increased, and the casualties is at a serious level. For this reason, the government is establishing and promoting policies to open traffic accident data and solve problems. In this paper, describe the method of predicting traffic accidents by eliminating the class imbalance using the traffic accident data and constructing the Hybrid Model. Using the original traffic accident data and the sampled data as learning data which use FP-Growth algorithm it learn patterns associated with traffic accident injury severity. Accordingly, In this paper purpose a method for predicting the severity of a victim of a traffic accident by analyzing the association patterns of two learning data, we can extract the same related patterns, when a decision tree and multinomial logistic regression analysis are performed, a hybrid model is constructed by assigning weights to related attributes.

• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.125-140
• /
• 2013

We call a data set in which the number of records belonging to a certain class far outnumbers the number of records belonging to the other class, 'imbalanced data set'. Most of the classification techniques perform poorly on imbalanced data sets. When we evaluate the performance of a certain classification technique, we need to measure not only 'accuracy' but also 'sensitivity' and 'specificity'. In a customer churn prediction problem, 'retention' records account for the majority class, and 'churn' records account for the minority class. Sensitivity measures the proportion of actual retentions which are correctly identified as such. Specificity measures the proportion of churns which are correctly identified as such. The poor performance of the classification techniques on imbalanced data sets is due to the low value of specificity. Many previous researches on imbalanced data sets employed 'oversampling' technique where members of the minority class are sampled more than those of the majority class in order to make a relatively balanced data set. When a classification model is constructed using this oversampled balanced data set, specificity can be improved but sensitivity will be decreased. In this research, we developed a hybrid model of support vector machine (SVM), artificial neural network (ANN) and decision tree, that improves specificity while maintaining sensitivity. We named this hybrid model 'hybrid SVM model.' The process of construction and prediction of our hybrid SVM model is as follows. By oversampling from the original imbalanced data set, a balanced data set is prepared. SVM_I model and ANN_I model are constructed using the imbalanced data set, and SVM_B model is constructed using the balanced data set. SVM_I model is superior in sensitivity and SVM_B model is superior in specificity. For a record on which both SVM_I model and SVM_B model make the same prediction, that prediction becomes the final solution. If they make different prediction, the final solution is determined by the discrimination rules obtained by ANN and decision tree. For a record on which SVM_I model and SVM_B model make different predictions, a decision tree model is constructed using ANN_I output value as input and actual retention or churn as target. We obtained the following two discrimination rules: 'IF ANN_I output value <0.285, THEN Final Solution = Retention' and 'IF ANN_I output value ${\geq}0.285$, THEN Final Solution = Churn.' The threshold 0.285 is the value optimized for the data used in this research. The result we present in this research is the structure or framework of our hybrid SVM model, not a specific threshold value such as 0.285. Therefore, the threshold value in the above discrimination rules can be changed to any value depending on the data. In order to evaluate the performance of our hybrid SVM model, we used the 'churn data set' in UCI Machine Learning Repository, that consists of 85% retention customers and 15% churn customers. Accuracy of the hybrid SVM model is 91.08% that is better than that of SVM_I model or SVM_B model. The points worth noticing here are its sensitivity, 95.02%, and specificity, 69.24%. The sensitivity of SVM_I model is 94.65%, and the specificity of SVM_B model is 67.00%. Therefore the hybrid SVM model developed in this research improves the specificity of SVM_B model while maintaining the sensitivity of SVM_I model.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.22 no.5
• /
• pp.391-401
• /
• 2009

The hybrid approach for structural experiments decomposes a structure into independent substructures that can be tested or simulated. The results from the decomposed substructures are combined to predict the behaviors of the entires structure. The hybrid approach is especially useful for the hybrid pseudo-dynamic tests that overcome the limitations of size of a test structure present in a shaking table test. The development of a computer system for the hybrid experiment requires a data model that formally organizes the information involved in the hybrid experiments. This paper provides the data model for representing the information involved in the hybrid experiments, by modifying the classes and attributes for the hybrid experiments in the Lehigh Model that is one of the data models for structural experiments. The data model for the hybrid experiments includes the classes for the physical substructures being tested and the analytical substructures being analyzed, and the simulation coordinator managing the overall experiments. Some objects for classes are implemented as an example to show the links among the classes. The data model presented in this paper can be applied for developing a computer system that helps structural engineers and researchers store, share, and access the information for the hybrid experiments.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.9 no.1
• /
• pp.8-19
• /
• 2001

A number of atomization and droplet breakup models have been developed and used to predict the diesel spray characteristic. Most of these models could not provide reasonable computational result of the diesel spray characteristic because they have only considered the primary breakup. A hybrid model is, therefore, required to develop by considering the primary and secondary breakup of liquid jet. according to this approach, wave breakup(WB) model was used compute the primary breakup of the liquid jet and droplet deformation and breakup(DDB) model was used for the secondary breakup of droplet. Development of hybrid model by using KIVA-II code was performed by comparing with the experimental data of spray tip penetration and SMD from the literature. A hybrid model developed in this study could provide the good agreement with the experimental data of spray tip penetration. The prediction results of SMD were in good agreement between 0.5 and 1.0 ms after the start of injection. Numerical results obtained by the present hybrid model have the good agreement with the experimental data with the breakup time constant in WB model of 30, and DDB model constant Ck of 1.0 when the droplet becomes less than 95% of maximum droplet diameter injected.

• Journal of the Korean Data and Information Science Society
• /
• v.14 no.3
• /
• pp.641-651
• /
• 2003

This study suggests the hybrid models for interest rate forecasting using structural changes (or change points). The basic concept of this proposed model is to obtain significant intervals caused by change points, to identify them as the change-point groups, and to reflect them in interest rate forecasting. The model is composed of three phases. The first phase is to detect successive structural changes in the U. S. Treasury bill rate dataset. The second phase is to forecast the change-point groups with data mining classifiers. The final phase is to forecast interest rates with backpropagation neural networks (BPN). Based on this structure, we propose three hybrid models in terms of data mining classifier: (1) multivariate discriminant analysis (MDA)-supported model, (2) case-based reasoning (CBR)-supported model, and (3) BPN-supported model. Subsequently, we compare these models with a neural network model alone and, in addition, determine which of three classifiers (MDA, CBR and BPN) can perform better. For interest rate forecasting, this study then examines the prediction ability of hybrid models to reflect the structural change.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.3
• /
• pp.1089-1098
• /
• 2018

This study aims at developing and applying a hybrid model to the wind power prediction (WPP). The hybrid model for a very-short-term WPP (VSTWPP) is achieved through analytical data, multiple linear regressions and least square methods (MLR&LS). The data used in our hybrid model are based on the historical records of wind power from an offshore region. In this model, the WPP is achieved in four steps: 1) transforming historical data into ratios; 2) predicting the wind power using the ratios; 3) predicting rectification ratios by the total wind power; 4) predicting the wind power using the proposed rectification method. The proposed method includes one-step and multi-step predictions. The WPP is tested by applying different models, such as the autoregressive moving average (ARMA), support vector machine (SVM), and artificial neural network (ANN). The results of all these models confirmed the validity of the proposed hybrid model in terms of error as well as its effectiveness. Furthermore, forecasting errors are compared to depict a highly variable WPP, and the correlations between the actual and predicted wind powers are shown. Simulations are carried out to definitely prove the feasibility and excellent performance of the proposed method for the VSTWPP versus that of the SVM, ANN and ARMA models.

• Journal of Korean Institute of Industrial Engineers
• /
• v.37 no.1
• /
• pp.74-82
• /
• 2011

The Bass model is a cornerstone in diffusion theory which is used for forecasting demand of durables or new services. Three well-known estimation methods for parameters of the Bass model are Ordinary Least Square (OLS), Maximum Likelihood Estimator (MLE), Nonlinear Least Square (NLS). In this paper, a hybrid method incorporating OLS and NLS is presented and it's performance is analyzed and compared with OLS and NLS by using simulation data and empirical data. The results show that NLS has the best performance in terms of accuracy and our hybrid method has the best performance in terms of stability. Specifically, hybrid method has better performance with less data. This result means much in practical aspect because the avaliable data is little when a diffusion model is used for forecasting demand of a new product.

• International Journal of Computer Science & Network Security
• /
• v.22 no.6
• /
• pp.310-318
• /
• 2022

Job offers have become more widespread and it has become easier and faster to apply for jobs through electronic recruitment platforms. In order to increase the protection of the data that is attached to the recruitment platforms. In this research, a proposed model was created through the use of hybrid encryption, which is used through the following algorithms: AES,Twofish,. This proposed model proved the effectiveness of using hybrid encryption in protecting personal data.