• Journal of IKEEE
• /
• v.27 no.3
• /
• pp.313-318
• /
• 2023

This paper proposes a neural network applying fine-tuning as a way to improve the performance of Video Classification based on Vision Transformer. Recently, the need for real-time video image analysis based on deep learning has emerged. Due to the characteristics of the existing CNN model used in Image Classification, it is difficult to analyze the association of consecutive frames. We want to find and solve the optimal model by comparing and analyzing the Vision Transformer and Non-local neural network models with the Attention mechanism. In addition, we propose an optimal fine-tuning neural network model by applying various methods of fine-tuning as a transfer learning method. The experiment trained the model with the UCF101 dataset and then verified the performance of the model by applying a transfer learning method to the UTA-RLDD dataset.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.8
• /
• pp.333-340
• /
• 2023

Efficient prediction of corporate bankruptcy is an important part of making appropriate lending decisions for financial institutions and reducing loan default rates. In many studies, classification models using artificial intelligence technology have been used. In the financial industry, even if the performance of the new predictive models is excellent, it should be accompanied by an intuitive explanation of the basis on which the result was determined. Recently, the US, EU, and South Korea have commonly presented the right to request explanations of algorithms, so transparency in the use of AI in the financial sector must be secured. In this paper, an artificial intelligence-based interpretable classification prediction model was proposed using corporate bankruptcy data that was open to the outside world. First, data preprocessing, 5-fold cross-validation, etc. were performed, and classification performance was compared through optimization of 10 supervised learning classification models such as logistic regression, SVM, XGBoost, and LightGBM. As a result, LightGBM was confirmed as the best performance model, and SHAP, an explainable artificial intelligence technique, was applied to provide a post-explanation of the bankruptcy prediction process.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.4
• /
• pp.222-227
• /
• 2022

In this paper, we implemented a Korean text generation and classification model based on a deep learning algorithm that can be applied to various industries. It consists of two implemented GAN-based Korean handwriting generation models and CNN-based Korean handwriting classification models. The GAN model consists of a generator model for generating fake Korean handwriting data and a discriminator model for discriminating fake handwritten data. In the case of the CNN model, the model was trained using the 'PHD08' dataset, and the learning result was 92.45. It was confirmed that Korean handwriting was classified with % accuracy. As a result of evaluating the performance of the classification model by integrating the Korean cursive data generated through the implemented GAN model and the training dataset of the existing CNN model, it was confirmed that the classification performance was 96.86%, which was superior to the existing classification performance.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.27-35
• /
• 2023

In this paper, we propose a process of increasing productivity by applying a deep learning-based defect detection and classification system to the prepreg fiber manufacturing process, which is in high demand in the field of producing composite materials. In order to apply it to toe prepreg manufacturing equipment that requires a solution due to the occurrence of a large amount of defects in various conditions, the optimal environment was first established by selecting cameras and lights necessary for defect detection and classification model production. In addition, data necessary for the production of multiple classification models were collected and labeled according to normal and defective conditions. The multi-classification model is made based on CNN and applies pre-learning models such as VGGNet, MobileNet, ResNet, etc. to compare performance and identify improvement directions with accuracy and loss graphs. Data augmentation and dropout techniques were applied to identify and improve overfitting problems as major problems. In order to evaluate the performance of the model, a performance evaluation was conducted using the confusion matrix as a performance indicator, and the performance of more than 99% was confirmed. In addition, it checks the classification results for images acquired in real time by applying them to the actual process to check whether the discrimination values are accurately derived.

• Journal of Information Technology Applications and Management
• /
• v.15 no.4
• /
• pp.21-36
• /
• 2008

As the great disturbance from abusing credit cards in Korea becomes stabilized, credit card companies need to interpret credit-card delinquents classification models from the viewpoint of profit. However, hit ratio which has been used as a measure of goodness of classification models just tells us how much correctly they classified rather than how much profits can be obtained as a result of using classification models. In this research, we tried to develop a new utility-based measure from the viewpoint of profit and then used this new measure to analyze two classification models(Neural Networks and Decision Tree models). We found that the hit ratio of neural model is higher than that of decision tree model, but the utility value of decision tree model is higher than that of neural model. This experiment shows the importance of utility based measure for credit-card delinquents classification models. We expect this new measure will contribute to increasing profits of credit card companies.

• Journal of Korea Multimedia Society
• /
• v.20 no.10
• /
• pp.1637-1645
• /
• 2017

Target/Nontarget classification can be divided into the study of shape estimation of the target analysing reflected echo signal and of type classification of the target using acoustical features. In active sonar system, the feature vectors are extracted from the signal reflected from the target, and an classification algorithm is applied to determine whether the received signal is a target or not. However, received sonar signals can be distorted in the underwater environments, and the spatio-temporal characteristics of active sonar signals change according to the aspect of the target. In addition, it is very difficult to collect real sea-trial data for research. In this paper, target/non-target classification were performed using real sea-trial data. Feature vectors are extracted using MFCC(Mel-Frequency Cepstral Coefficients), filterbank energy in the Fourier spectrum and wavelet domain. For the performance verification, classification experiments were performed using backpropagation neural network classifiers.

• ETRI Journal
• /
• v.33 no.6
• /
• pp.871-879
• /
• 2011

Because a wide variety of multimedia services are provided through personal wireless communication devices, the demand for efficient bandwidth utilization becomes stronger. This demand naturally results in the introduction of the variable bitrate speech coding concept. One exemplary work is the selectable mode vocoder (SMV) that supports speech/music classification. However, because it has severe limitations in its classification performance, a couple of works to improve speech/music classification by introducing support vector machines (SVMs) have been proposed. While these approaches significantly improved classification accuracy, they did not consider correlations commonly found in speech and music frames. In this paper, we propose a novel and orthogonal approach to improve the speech/music classification of SMV codec by adaptively tuning SVMs based on interframe correlations. According to the experimental results, the proposed algorithm yields improved results in classifying speech and music within the SMV framework.

• Genomics & Informatics
• /
• v.2 no.1
• /
• pp.19-29
• /
• 2004

Gene expression profiles may offer more information than morphology and provide an alternative to morphology- based tumor classification systems. Informative gene selection is finding gene subsets that are able to discriminate between tumor types, and may have clear biological interpretation. Gene selection is a fundamental issue in gene expression based tumor classification. In this report, techniques for selecting informative genes are illustrated and supervised shaving introduced as a gene selection method in the place of a clustering algorithm. The supervised shaving method showed good performance in gene selection and classification, even though it is a clustering algorithm. Almost selected genes are related to leukemia disease. The expression profiles of 3051 genes were analyzed in 27 acute lymphoblastic leukemia and 11 myeloid leukemia samples. Through these examples, the supervised shaving method has been shown to produce biologically significant genes of more than $94\%$ accuracy of classification. In this report, SVM has also been shown to be a practicable method for gene expression-based classification.

• Journal of The Korean Association For Science Education
• /
• v.32 no.6
• /
• pp.1063-1071
• /
• 2012

The purpose of this study is to investigate the relation between the classification ability quotient and cortisol-hormonal change of middle school students. Thirty-three students, second graders in middle school, performed the classification task that can be an indicator of students' classification ability. And then amount of the secreted hormone was analyzed during task performance. The study results were as follows: First, the classification methods of students mostly utilized visual, qualitative. Their classification patterns for each subject were static, partial, and non-comparative. Second, the amount of stress-hormone was secreted from students during the experiment decreased in overall after the free classification. It seemed that student-centered activity relieved stress. Third, the classification ability quotient turned out to be significantly correlated to the stress hormone, which means that there was a close relationship between classification ability and stress level. It was also considered that stress had a positive effect on the improvement of classification ability. This study provided physiologically more accurate information on the stress increased in the learning process than other conventional studies based on reports or interviews. Finally, researchers could recognize the effect of stress in the cognitive activity and the need to find an appropriate level of stress in learning processes.

• The Korean Journal of Ecology
• /
• v.26 no.2
• /
• pp.75-81
• /
• 2003

This study was performed to prove if the high resolution satellite imagery of IKONOS is suitable for preparing digital vegetation map which is becoming increasingly important in ecological science. Seven classes for forest area and five classes for non-forest area were taken for classification. Three methods, such as the pixel based classification, the segment based classification with majority principle, and the segment based classification with maximum likelihood, were applied to classify IKONOS imagery taken in April 2000. As a whole, the segment based classification shows better performance in classifying the high resolution satellite imagery of IKONOS. Through the comparison of accuracies and kappa values of the above 3 classification methods, the segment based classification with maximum likelihood was proved to be the best suitable for preparing the vegetation map with the help of IKONOS imagery. This is true not only from the viewpoint of accuracy, but also for the purpose of preparing a polygon based vegetation map. On the basis of the segment based classification with the maximum likelihood, a digital vegetation map in which each vegetation class is delimitated in the form of a polygon could be prepared.