• The Journal of Engineering Geology
• /
• v.30 no.4
• /
• pp.457-468
• /
• 2020

Problems arising during pile design works for plant construction, civil and architecture work are mostly come from uncertainty of geotechnical characteristics. In particular, obtaining the N-value measured through the Standard Penetration Test (SPT) is the most important data. However, it is difficult to obtain N-value by drilling investigation throughout the all target area. There are many constraints such as licensing, time, cost, equipment access and residential complaints etc. it is impossible to obtain geotechnical characteristics through drilling investigation within a short bidding period in overseas. The geotechnical characteristics at non-drilling investigation points are usually determined by the engineer's empirical judgment, which can leads to errors in pile design and quantity calculation causing construction delay and cost increase. It would be possible to overcome this problem if N-value could be predicted at the non-drilling investigation points using limited minimum drilling investigation data. This study was conducted to predicted the N-value using an Artificial Neural Network (ANN) which one of the Artificial intelligence (AI) method. An Artificial Neural Network treats a limited amount of geotechnical characteristics as a biological logic process, providing more reliable results for input variables. The purpose of this study is to predict N-value at the non-drilling investigation points through patterns which is studied by multi-layer perceptron and error back-propagation algorithms using the minimum geotechnical data. It has been reviewed the reliability of the values that predicted by AI method compared to the measured values, and we were able to confirm the high reliability as a result. To solving geotechnical uncertainty, we will perform sensitivity analysis of input variables to increase learning effect in next steps and it may need some technical update of program. We hope that our study will be helpful to design works in the future.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.221-241
• /
• 2018

Deep learning is getting attention recently. The deep learning technique which had been applied in competitions of the International Conference on Image Recognition Technology(ILSVR) and AlphaGo is Convolution Neural Network(CNN). CNN is characterized in that the input image is divided into small sections to recognize the partial features and combine them to recognize as a whole. Deep learning technologies are expected to bring a lot of changes in our lives, but until now, its applications have been limited to image recognition and natural language processing. The use of deep learning techniques for business problems is still an early research stage. If their performance is proved, they can be applied to traditional business problems such as future marketing response prediction, fraud transaction detection, bankruptcy prediction, and so on. So, it is a very meaningful experiment to diagnose the possibility of solving business problems using deep learning technologies based on the case of online shopping companies which have big data, are relatively easy to identify customer behavior and has high utilization values. Especially, in online shopping companies, the competition environment is rapidly changing and becoming more intense. Therefore, analysis of customer behavior for maximizing profit is becoming more and more important for online shopping companies. In this study, we propose 'CNN model of Heterogeneous Information Integration' using CNN as a way to improve the predictive power of customer behavior in online shopping enterprises. In order to propose a model that optimizes the performance, which is a model that learns from the convolution neural network of the multi-layer perceptron structure by combining structured and unstructured information, this model uses 'heterogeneous information integration', 'unstructured information vector conversion', 'multi-layer perceptron design', and evaluate the performance of each architecture, and confirm the proposed model based on the results. In addition, the target variables for predicting customer behavior are defined as six binary classification problems: re-purchaser, churn, frequent shopper, frequent refund shopper, high amount shopper, high discount shopper. In order to verify the usefulness of the proposed model, we conducted experiments using actual data of domestic specific online shopping company. This experiment uses actual transactions, customers, and VOC data of specific online shopping company in Korea. Data extraction criteria are defined for 47,947 customers who registered at least one VOC in January 2011 (1 month). The customer profiles of these customers, as well as a total of 19 months of trading data from September 2010 to March 2012, and VOCs posted for a month are used. The experiment of this study is divided into two stages. In the first step, we evaluate three architectures that affect the performance of the proposed model and select optimal parameters. We evaluate the performance with the proposed model. Experimental results show that the proposed model, which combines both structured and unstructured information, is superior compared to NBC(Naïve Bayes classification), SVM(Support vector machine), and ANN(Artificial neural network). Therefore, it is significant that the use of unstructured information contributes to predict customer behavior, and that CNN can be applied to solve business problems as well as image recognition and natural language processing problems. It can be confirmed through experiments that CNN is more effective in understanding and interpreting the meaning of context in text VOC data. And it is significant that the empirical research based on the actual data of the e-commerce company can extract very meaningful information from the VOC data written in the text format directly by the customer in the prediction of the customer behavior. Finally, through various experiments, it is possible to say that the proposed model provides useful information for the future research related to the parameter selection and its performance.

• Korean Journal of Remote Sensing
• /
• v.28 no.5
• /
• pp.489-499
• /
• 2012

In this paper, a classifier ensemble framework for remote sensing data classification is presented that combines classification results generated from both different training sets and different classifiers. A core part of the presented framework is to increase a diversity between classification results by using both different training sets and classifiers to improve classification accuracy. First, different training sets that have different sampling densities are generated and used as inputs for supervised classification using different classifiers that show different discrimination capabilities. Then several preliminary classification results are combined via a majority voting scheme to generate a final classification result. A case study of land-cover classification using multi-temporal ENVISAT ASAR data sets is carried out to illustrate the potential of the presented classification framework. In the case study, nine classification results were combined that were generated by using three different training sets and three different classifiers including maximum likelihood classifier, multi-layer perceptron classifier, and support vector machine. The case study results showed that complementary information on the discrimination of land-cover classes of interest would be extracted within the proposed framework and the best classification accuracy was obtained. When comparing different combinations, to combine any classification results where the diversity of the classifiers is not great didn't show an improvement of classification accuracy. Thus, it is recommended to ensure the greater diversity between classifiers in the design of multiple classifier systems.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.2 no.2
• /
• pp.69-78
• /
• 1999

In this paper, a neural network approach to forecast Korean regional precipitation is presented. We first analyze the characteristics of the conventional models for time series prediction, and then propose a new model and its learning method for the precipitation forecast. The proposed model is a layered network in which the outputs of a layer are buffered within a given period time and then fed fully connected to the upper layer. This study adopted the dual connections between two layers for the model. The network behavior and learning algorithm for the model are also described. The dual connection structure plays the role of the bias of the ordinary Multi-Layer Perceptron(MLP), and reflects the relationships among the features effectively. From these advantageous features, the model provides the learning efficiency in comparison with the FIR network, which is the most popular model for time series prediction. We have applied the model to the monthly and seasonal forecast of precipitation. The precipitation data and SST(Sea Surface Temperature) data for several decades are used as the learning pattern for the neural network predictor. The experimental results have shown the validity of the proposed model.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.3
• /
• pp.107-115
• /
• 2009

Skin color detection is used to classify input pixels into skin and non skin area, and it requires the classifier to have a high classification rate. In previous work, most classifiers used single color model for skin color detection. However the classification rate can be increased by using more than one color model due to the various characteristics of skin color distribution in different color models, and the MLP is also invested as a more efficient classifier with less parameters than other classifiers. But the input dimension and required parameters of MLP will be increased when using two color models in skin color detection, as a result, the increased parameters will cause the huge teaming time in MLP. In this paper, we propose a MLP based classifier with less parameters in two color models. The proposed partially connected MLP based on two color models can reduce the number of weights and improve the classification rate. Because the characteristic of different color model can be learned in different partial networks. As the experimental results, we obtained 91.8% classification rate when testing various images in RGB and CbCr models.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.6
• /
• pp.57-65
• /
• 2010

In this paper, we present a universal steganalysis scheme. The proposed method extract features of two types. First feature set is extracted from histogram characteristic of the wavelet subbands. Second feature set is determined by statistical moments of wavelet characteristic functions. 3-level wavelet decomposition is performed for stego image and cover image using the Haar wavelet basis. We extract one features from 9 high frequency subbands of 12 subbands. The number of second features is 39. We use total 48 features for steganalysis. Multi layer perceptron(MLP) is applied as classifier to distinguish between cover images and stego images. To evaluate the proposed steganalysis method, we use the CorelDraw image database. We test the performance of our proposed steganalysis method over LSB method, spread spectrum data hiding method, blind spread spectrum data hiding method and F5 data hiding method. The proposed method outperforms the previous methods in sensitivity, specificity, error rate and area under ROC curve, etc.

• Journal of Korea Water Resources Association
• /
• v.44 no.2
• /
• pp.97-107
• /
• 2011

For the purpose of enhancing usability of NWP (Numerical Weather Prediction), the quantitative precipitation prediction scheme was suggested. In this research, precipitation by leading time was predicted using 3-hour rainfall accumulation by meso-scale numerical weather model and AWS (Automatic Weather Station), precipitation water and relative humidity observed by atmospheric sounding station, probability of rainfall occurrence by leading time in June and July, 2001 and August, 2002. Considering the nonlinear process of ranfall producing mechanism, the ANN (Artificial Neural Network) that is useful in nonlinear fitting between rainfall and the other atmospheric variables. The feedforward multi-layer perceptron was used for neural network structure, and the nonlinear bipolaractivation function was used for neural network training for converting negative rainfall into no rain value. The ANN simulated rainfall was validated by leading time using Nash-Sutcliffe Coefficient of Efficiency (COE) and Coefficient of Correlation (CORR). As a result, the 3 hour rainfall accumulation basis shows that the COE of the areal mean of the Korean peninsula was improved from -0.04 to 0.31 for the 12 hr leading time, -0.04 to 0.38 for the 24 hr leading time, -0.03 to 0.33 for the 36 hr leading time, and -0.05 to 0.27 for the 48 hr leading time.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.42 no.1
• /
• pp.9-26
• /
• 2005

In the pursuit of ever higher levels of performance, recent computer systems have made use of deep pipeline, dynamic scheduling and multi-issue superscalar processor technologies. In this situations, branch prediction schemes are an essential part of modem microarchitectures because the penalty for a branch misprediction increases as pipelines deepen and the number of instructions issued per cycle increases. In this paper, we propose a novel branch prediction scheme, direction-gshare(d-gshare), to improve the prediction accuracy. At first, we model a neural network with the components that possibly affect the branch prediction accuracy, and analyze the variation of their weights based on the neural network information. Then, we newly add the component that has a high weight value to an original gshare scheme. We simulate our branch prediction scheme using Simple Scalar, a powerful event-driven simulator, and analyze the simulation results. Our results show that, compared to bimodal, two-level adaptive and gshare predictor, direction-gshare predictor(d-gshare. 3) outperforms, without additional hardware costs, by up to 4.1% and 1.5% in average for the default mont of embedded direction, and 11.8% in maximum and 3.7% in average for the optimal one.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.4
• /
• pp.93-101
• /
• 2004

A face boundary can be approximated by an ellipse with five-dimensional parameters. This property allows an ellipse detection algorithm to be adapted to detecting faces. However, the construction of a huge five-dimensional parameter space for a Hough transform is quite unpractical. Accordingly, we Propose a selectively attentional Hough transform method for detecting faces from a symmetric contour in an image. The idea is based on the use of a constant aspect ratio for a face, gradient information, and scan-line-based orientation decomposition, thereby allowing a 5-dimensional problem to be decomposed into a two-dimensional one to compute a center with a specific orientation and an one-dimensional one to estimate a short axis. In addition, a two-point selection constraint using geometric and gradient information is also employed to increase the speed and cope with a cluttered background. After detecting candidate face regions using the proposed Hough transform, a multi-layer perceptron verifier is adopted to reject false positives. The proposed method was found to be relatively fast and promising.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.3 no.3
• /
• pp.9-20
• /
• 1999

A method is proposed to estimate the joint damages of a steel structure from modal data using the neural networks technique. The beam-to-column connection in a steel frame structure is represented by a zero-length rotational spring of the end of the beam element, and the connection fixity factor is defined based on the rotational stiffness so that the factor may be in the range 0~1.0. Then, the severity of joint damage is defined as the reduction ratio of the connection fixity factor. Several advanced techniques are employed to develop the robust damage identification technique using neural networks. The concept of the substructural indentification is used for the localized damage assessment in the large structure. The noise-injection learning algorithm is used to reduce the effects of the noise in the modal data. The data perturbation scheme is also employed to assess the confidence in the estimated damages based on a few sets of actual measurement data. The feasibility of the proposed method is examined through a numerical simulation study on a 2-bay 10-story structure and an experimental study on a 2-story structure. It has been found that the joint damages can be reasonably estimated even for the case where the measured modal vectors are limited to a localized substructure and the data are severely corrupted with noise.