• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 2001.10a
• /
• pp.59-64
• /
• 2001

The objectives of the thesis are to propose a pattern classification method for remote sensing data using artificial neural network. First, we apply the error back propagation algorithm to classify the remote sensing data. In this case, the classification performance depends on a training data set. Using the training data set and the error back propagation algorithm, a layered neural network is trained such that the training pattern are classified with a specified accuracy. After training the neural network, some pixels are deleted from the original training data set if they are incorrectly classified and a new training data set is built up. Once training is complete, a testing data set is classified by using the trained neural network. The classification results of Landsat TM data show that this approach produces excellent results which are more realistic and noiseless compared with a conventional Bayesian method.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.151-152
• /
• 2019

In this paper, we propose an improved learning method based on a small data set for animal image classification. First, CNN creates a training model for a small data set and uses the data set to expand the data set of the training set Second, a bottleneck of a small data set is extracted using a pre-trained network for a large data set such as VGG16 and stored in two NumPy files as a new training data set and a test data set, finally, learn the fully connected network as a new data set.

• Convergence Security Journal
• /
• v.20 no.4
• /
• pp.71-80
• /
• 2020

In order to provide realistic traffic to the cyber warfare training system, it is necessary to prepare a traffic distribution plan in advance and to create a training data set using normal/threat data sets. This paper presents the design and implementation results of a method for creating a traffic distribution plan and a training data set to provide background traffic like a real environment to a cyber warfare training system. We propose a method of a traffic distribution plan by using the network topology of the training environment to distribute traffic and the traffic attribute information collected in real and simulated environments. We propose a method of generating a training data set according to a traffic distribution plan using a unit traffic and a mixed traffic method using the ratio of the protocol. Using the implemented tool, a traffic distribution plan was created, and the training data set creation result according to the distribution plan was confirmed.

• Journal of IKEEE
• /
• v.22 no.3
• /
• pp.816-821
• /
• 2018

Recently, machine learning algorithms based on artificial neural networks started to be used widely as classifiers in the field of EEG research for emotion analysis and disease diagnosis. When a machine learning model is used to classify EEG data, if training data is composed of only data having similar characteristics, classification performance may be deteriorated when applied to data of another group. In this paper, we propose a method to construct training data set by selecting several groups of data using semi-supervised learning algorithm to improve these problems. We then compared the performance of the two models by training the model with a training data set consisting of data with similar characteristics to the training data set constructed using the proposed method.

• The Journal of Information Technology and Database
• /
• v.2 no.2
• /
• pp.71-85
• /
• 1995

Many research results conducted by neural network researchers have claimed that the classification accuracy of neural networks is superior to, or at least equal to that of conventional methods. However, in series of neural network classifications, it was found that the classification accuracy strongly depends on the characteristics of training data set. Even though there are many research reports that the classification accuracy of neural networks can be different, depending on the composition and architecture of the networks, training algorithm, and test data set, very few research addressed the problem of classification accuracy when the basic assumption of data monotonicity is violated, In this research, development project of automated credit evaluation system is described. The finding was that arrangement of training data is critical to successful implementation of neural training to maintain monotonicity of the data set, for enhancing classification accuracy of neural networks.

• Journal of the Korea Convergence Society
• /
• v.8 no.2
• /
• pp.27-33
• /
• 2017

To efficiently provide cloud computing services to users over the Internet, IT resources must be configured in the data center based on virtualization and distributed computing technology. This paper focuses specifically on the problem that new training data can be added at any time in a wide range of fields, and new attributes can be added to training data at any time. In such a case, rule generated by the training data with the former attribute set can not be used. Moreover, the rule can not be combined with the new data set(with the newly added attributes). This paper proposes further development of the new inference engine that can handle the above case naturally. Rule generated from former data set can be combined with the new data set to form the refined rule.

• Korean Journal of Remote Sensing
• /
• v.18 no.6
• /
• pp.369-376
• /
• 2002

This paper presents a method to enhance the performance of supervised classification by separating the spectral signature of the training data sets for each class. Using clustering technique, a training data set is divided into several subsets which show a pattern of the normal distribution with small value of spectral variances. Then a supervised classification is applied with the divided training data set as training data for the temporary subclasses of the original class. The proposed method is applied to a Landsat TM image of Busan area for the applicability test. The result shows that the proposed method produces better classified results than the conventional statistical classification methods. It is expected that the proposed method will reduce the effort and expense for selecting the training data set for each class in an area which has spectrally homogeneous signature.

• Journal of Intelligence and Information Systems
• /
• v.5 no.1
• /
• pp.95-101
• /
• 1999

In this paper, a comparison between different methods to combine predictions from neural networks will be given. These methods are bagging, bumping, and balancing. Those are based on the analysis of the ensemble generalization error into an ambiguity term and a term incorporating generalization performances of individual networks. Neural Networks and AI machine learning models are prone to overfitting. A strategy to prevent a neural network from overfitting, is to stop training in early stage of the learning process. The complete data set is spilt up into a training set and a validation set. Training is stopped when the error on the validation set starts increasing. The stability of the networks is highly dependent on the division in training and validation set, and also on the random initial weights and the chosen minimization procedure. This causes early stopped networks to be rather unstable: a small change in the data or different initial conditions can produce large changes in the prediction. Therefore, it is advisable to apply the same procedure several times starting from different initial weights. This technique is often referred to as training ensembles of neural networks. In this paper, we presented a comparison of three statistical methods to prevent overfitting of neural network.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.88-90
• /
• 2022

Semi Supervised Learning 은 일부의 data 에는 labeling 을 하고 나머지 data 에는 labeling 을 안한채로 학습을 진행하는 방법이다. Object Detection 은 이미지에서 여러개의 객체들의 대한 위치를 여러개의 바운딩 박스로 지정해서 찾는 Computer Vision task 이다. 당연하게도, model training 단계에서 사용되는 data set 의 크기가 크고 객체가 많을 수록 일반적으로 model 의 성능이 좋아 질 것이다. 하지만 실험 환경에 따라 data set 을 잘 확보하지 못하던가, 실험 장치가 데이터 셋을 감당하지 못하는 등의 문제가 발생 할 수 있다. 그렇기에 본 논문에서는 semi supervised learning based object detection model 을 알아보고 data set 의 크기를 조절해가며 modle 을 training 시킨 뒤 data set 의 크기에 따라 성능이 어떻게 변화하는 지를 알아 볼 것이다.

• 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.