• Korean Journal of Remote Sensing
• /
• v.18 no.3
• /
• pp.155-162
• /
• 2002

Neural networks have been focused on as a robust classifier for the remotely sensed imagery due to its statistical independency and teaming ability. Also the artificial neural networks have been reported to be more tolerant to noise and missing data. However, unlike the conventional statistical classifiers which use the statistical parameters for the classification, a neural network classifier uses individual training sample in teaming stage. The training performance of a neural network is know to be very sensitive to the discrepancy of the number of the training samples of each class. In this paper, the effect of the population discrepancy of training samples of each class was analyzed with three layered feed forward network. And a method for reducing the effect was proposed and experimented with Landsat TM image. The results showed that the effect of the training sample size discrepancy should be carefully considered for faster and more accurate training of the network. Also, it was found that the proposed method which makes teaming rate as a function of the number of training samples in each class resulted in faster and more accurate training of the network.

• Journal of Broadcast Engineering
• /
• v.24 no.3
• /
• pp.472-484
• /
• 2019

This paper proposes a time-domain sound event detection algorithm using DNN (Deep Neural Network). In this system, time domain sound waveform data which is not converted into the frequency domain is used as input to the DNN. The overall structure uses CRNN structure, and GLU, ResNet, and Squeeze-and-excitation blocks are applied. And proposed structure uses structure that considers features extracted from several layers together. In addition, under the assumption that it is practically difficult to obtain training data with strong labels, this study conducted training using a small number of weakly labeled training data and a large number of unlabeled training data. To efficiently use a small number of training data, the training data applied data augmentation methods such as time stretching, pitch change, DRC (dynamic range compression), and block mixing. Unlabeled data was supplemented with insufficient training data by attaching a pseudo-label. In the case of using the neural network and the data augmentation method proposed in this paper, the sound event detection performance is improved by about 6 %(based on the f-score), compared with the case where the neural network of the CRNN structure is used by training in the conventional method.

• Journal of Korea Multimedia Society
• /
• v.24 no.9
• /
• pp.1242-1250
• /
• 2021

Detection of outliers deviating normal data distribution in high dimensional data is an important technique in many application areas. In this paper, a distance-based outlier detection method using landmarks in high dimensional data is proposed. Given normal training data, the k-means clustering method is applied for the training data in order to extract the centers of the clusters as landmarks which represent normal data distribution. For a test data sample, the distance to the nearest landmark gives the outlier score. In the experiments using high dimensional data such as images and documents, it was shown that the proposed method based on the landmarks of one-tenth of training data can give the comparable outlier detection performance while reducing the time complexity greatly in the testing stage.

• Geophysics and Geophysical Exploration
• /
• v.24 no.3
• /
• pp.89-97
• /
• 2021

Fault detection in seismic data is well suited to the application of machine learning algorithms. Accordingly, various machine learning techniques are being developed. In recent studies, machine learning models, which utilize synthetic data, are the particular focus when training with deep learning. The use of synthetic training data has many advantages; Securing massive data for training becomes easy and generating exact fault labels is possible with the help of synthetic training data. To interpret real data with the model trained by synthetic data, the synthetic data used for training should be geologically realistic. In this study, we introduce a method to generate realistic synthetic seismic data. Initially, reflectivity models are generated to include realistic fault structures, and then, a one-way wave equation is applied to efficiently generate seismic stack sections. Next, a migration algorithm is used to remove diffraction artifacts and random noise is added to mimic actual field data. A convolutional neural network model based on the U-Net structure is used to verify the generated synthetic data set. From the results of the experiment, we confirm that realistic synthetic data effectively creates a deep learning model that can be applied to field data.

• Journal of Biosystems Engineering
• /
• v.40 no.3
• /
• pp.284-288
• /
• 2015

Training power tiller operators in safe farming is necessary to avoid farming accidents. With the continuing progress in computational technology, driving simulators have become increasingly popular for conducting such training. Purpose: The objective of this study is to develop mock control devices and data acquisition apparatus for a tiller simulator. Methods: Except for the stand and tail wheel adjusting levers, the mock control devices were developed using a tiller handle assay. The data acquisition apparatus was realized using an embedded data-logging device and LabVIEW, the system design software. Results: The control devices of a real handle assay were successfully mimicked by the mock operator control devices, which used sensors for the relevant measurements. The data from the mock devices were acquired and transmitted to the main computer at intervals of 10 ms via Wi-Fi. Conclusions: The developed mock control devices operate similar to real power tillers and can be utilized in power tiller training simulators.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.39 no.2
• /
• pp.111-130
• /
• 2014

For the key factors determining victory of combat, many works have been focusing on qualitative analyses in the past. As military training paradigm changes along with technology developments, demands for scientific analysis to prepare future military strength increase regarding military training results, and big data analysis has opened such possibility. We analyze the data from KCTC (Korea Combat Training Center) training to investigate the factors affected victory in offensive operations. In this context, we develop a way to measure the victory and the factors related to it from existing studies and military doctrines. We first identify Independent variables that affect offensive operations through variable selection and propose a mathematical model to explain combat victory by performing multiple regression analysis. We also verify our results with battalion-level live training data as well as previous studies on victory factors in the military doctrines.

• Nuclear Engineering and Technology
• /
• v.39 no.4
• /
• pp.337-348
• /
• 2007

A fuzzy neural network model is presented to predict residual stress for dissimilar metal welding under various welding conditions. The fuzzy neural network model, which consists of a fuzzy inference system and a neuronal training system, is optimized by a hybrid learning method that combines a genetic algorithm to optimize the membership function parameters and a least squares method to solve the consequent parameters. The data of finite element analysis are divided into four data groups, which are split according to two end-section constraints and two prediction paths. Four fuzzy neural network models were therefore applied to the numerical data obtained from the finite element analysis for the two end-section constraints and the two prediction paths. The fuzzy neural network models were trained with the aid of a data set prepared for training (training data), optimized by means of an optimization data set and verified by means of a test data set that was different (independent) from the training data and the optimization data. The accuracy of fuzzy neural network models is known to be sufficiently accurate for use in an integrity evaluation by predicting the residual stress of dissimilar metal welding zones.

• Journal of Practical Engineering Education
• /
• v.15 no.2
• /
• pp.381-388
• /
• 2023

Through artificial neural network education using spreadsheets, non-major undergraduate students can understand the operation principle of artificial neural networks and develop their own artificial neural network software. Here, training of the operation principle of artificial neural networks starts with the generation of training data and the assignment of correct answer labels. Then, the output value calculated from the firing and activation function of the artificial neuron, the parameters of the input layer, hidden layer, and output layer is learned. Finally, learning the process of calculating the error between the correct label of each initially defined training data and the output value calculated by the artificial neural network, and learning the process of calculating the parameters of the input layer, hidden layer, and output layer that minimize the total sum of squared errors. Training on the operation principles of artificial neural networks using a spreadsheet was conducted for undergraduate non-major students. And image training data and basic artificial neural network development results were collected. In this paper, we analyzed the results of collecting two types of training data and the corresponding artificial neural network SW with small 12-pixel images, and presented methods and execution results of using the collected training data for Orange machine learning model learning and analysis tools.

• Journal of Engineering Education Research
• /
• v.18 no.4
• /
• pp.3-12
• /
• 2015

The purpose of this study was to develop a pre-training program of engineering college students for maximizing the effectiveness of internship or field training. To pursue this goal, literature review was conducted for data collection about college and corporate pre-training program for internship or field training and pre-training program(draft) was proposed. A questionnaire survey was conducted with engineering professors, students and graduates to identify the needs for pre-training program(draft) for internship or field training. Based on the results, the contents of pre-training program for internship or field training were composed of basic liberal arts, basic competency, real information related with corporation or job, and information exchange network. And key consideration for operating the pre-training program for internship or field training were proposed with the management department, regulation for the obligatory participation, meaningful organizing content, feedback of needs.

• Research in Community and Public Health Nursing
• /
• v.13 no.2
• /
• pp.239-248
• /
• 2002

Objectives: The purpose of this study was to evaluate the effects of assertiveness training on assertiveness behavior, problem solving ability, and interpersonal relationships of nursing college students. Methods: A nonequivalent pre- and post-test experimental design was used. This study included 15 subjects in the experimental group, and 15 subjects in the control group, who were sophomore nursing students recruited from Gachon Nursing School located in Inchon City. The experimental group received 8 series of a ready planned and reorganized assertiveness training course, while the control group did not receive any training. The 120 minute training session was held on a weekly basis. The Maan-Whitney U Test was done to identify the changes in scores of assertiveness behavior, problem solving ability, and interpersonal relationships between the experimental and the control groups, after the assertiveness training. The data were collected before and after each part of the assertiveness training. The data collection period was from May 7 to July 10, 2001. Results: 1) The assertiveness behavior scores of the nursing students who participated in the assertiveness training were higher than those of the nursing students who did not participate in the assertiveness training. 2) The self problem solving evaluation scores of the nursing students who received assertiveness training were higher than those of the nursing students who did not receive the training. 3) The interpersonal relationship scores of the nursing students who participated in the assertiveness training were higher than those of the nursing students who did not participate in the assertiveness training. Conclusion: The study results showed that the assertiveness training was effective in promoting assertiveness behavior, problem solving ability, and changes in interpersonal relationships in nursing college students. It is suggested that well designed strategies are needed in the further studies in order to expand and apply the assertiveness training to other nursing student and nurse cohorts.