• Annual Conference of KIPS
• /
• 2004.05a
• /
• pp.1007-1010
• /
• 2004

본 논문에서는 분산 서비스거부 공격(DDoS)이 발생할 때 네트워크 트래픽의 특성을 분석하기 위해서 트래픽 비율분석법(TRA: Traffic Rate Analysis)을 제안하고 트래픽 비율분석법을 통해서 분석된 다양한 유형의 DDoS 공격의 특성을 기계학습(Machine Learning)을 이용해서 DDoS 공격의 탐지규칙을 생성하고 그 성능을 측정하였다. 트래픽 비율분석법은 감시대상 네트워크 트래픽에서 특정한 유형의 트래픽의 발생비율을 나타내며 TCP flag rate 와 Protocol rate 로 구분된다. 트래픽 비율분석법을 적용한 결과 각각의 DDoS 공격 유형에 따라서 매우 독특한 특성을 가짐을 발견하였다. 그리고, 분석된 데이터를 대상으로 세 개의 기계학습 방법(C4.5, CN2, Na?ve Bayesian Classifier)을 이용해서 DDoS 공격의 탐지규칙을 생성하여 DDoS 공격의 탐지에 적용했다. 실험결과, 본 논문에서 제안된 트래픽 비율분석법과 기계학습을 통한 DDoS 공격의 탐지방법은 매우 높은 수준의 성능을 나타냈다.

• Journal of information and communication convergence engineering
• /
• v.17 no.1
• /
• pp.41-48
• /
• 2019

In recent years, sentiment analysis research has become popular. The research results of sentiment analysis have achieved remarkable results in practical applications, such as in Amazon's book recommendation system and the North American movie box office evaluation system. Analyzing big data based on user preferences and evaluations and recommending hot-selling books and hot-rated movies to users in a targeted manner greatly improve book sales and attendance rate in movies [1, 2]. However, traditional machine learning-based sentiment analysis methods such as the Classification and Regression Tree (CART), Support Vector Machine (SVM), and k-nearest neighbor classification (kNN) had performed poorly in accuracy. In this paper, an improved kNN classification method is proposed. Through the improved method and normalizing of data, the purpose of improving accuracy is achieved. Subsequently, the three classification algorithms and the improved algorithm were compared based on experimental data. Experiments show that the improved method performs best in the kNN classification method, with an accuracy rate of 11.5% and a precision rate of 20.3%.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.438-440
• /
• 2021

With the development of machine learning, artificial intelligence techniques are being applied in various fields. Among these fields, there is an OCR technique that converts characters in images into text. The tesseract developed by HP is one of those techniques. However, the recognition rate for recognizing characters in images is still low. To this end, we try to improve the conversion rate of the text of the image through the post-processing process that recognizes the context using the ontology.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.2
• /
• pp.87-96
• /
• 2023

Recently, research on predicting the treatment results of diseases using deep learning technology is also active in the medical community. However, small patient data and specific deep learning algorithms were selected and utilized, and research was conducted to show meaningful results under specific conditions. In this study, in order to generalize the research results, patients were further expanded and subdivided to derive the results of a study predicting mortality after lung cancer diagnosis for men and women in their 80s, 90s, and 100s. Using AutoML, which provides large-scale medical information and various deep learning algorithms from the Health Insurance Review and Assessment Service, five algorithms such as Decision Tree, Random Forest, Gradient Boosting, XGBoost, and Logistic Registration were created to predict mortality rates for 84 months after lung cancer diagnosis. As a result of the study, men in their 80s and 90s had a higher mortality prediction rate than women, and women in their 100s had a higher mortality prediction rate than men. And the factor that has the greatest influence on the mortality rate was analyzed as the treatment period.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_4
• /
• pp.1111-1123
• /
• 2023

In recent times, semantic image segmentation methods using deep learning models have been widely used for monitoring changes in surface attributes using remote sensing imagery. To enhance the performance of various UNet-based deep learning models, including the prominent UNet model, it is imperative to have a sufficiently large training dataset. However, enlarging the training dataset not only escalates the hardware requirements for processing but also significantly increases the time required for training. To address these issues, transfer learning is used as an effective approach, enabling performance improvement of models even in the absence of massive training datasets. In this paper we present three transfer learning models, UNet-ResNet50, UNet-VGG19, and CBAM-DRUNet-VGG19, which are combined with the representative pretrained models of VGG19 model and ResNet50 model. We applied these models to building extraction tasks and analyzed the accuracy improvements resulting from the application of transfer learning. Considering the substantial impact of learning rate on the performance of deep learning models, we also analyzed performance variations of each model based on different learning rate settings. We employed three datasets, namely Kompsat-3A dataset, WHU dataset, and INRIA dataset for evaluating the performance of building extraction results. The average accuracy improvements for the three dataset types, in comparison to the UNet model, were 5.1% for the UNet-ResNet50 model, while both UNet-VGG19 and CBAM-DRUNet-VGG19 models achieved a 7.2% improvement.

• Journal of Practical Engineering Education
• /
• v.5 no.2
• /
• pp.163-168
• /
• 2013

The vicious cycle of over-education has been made. That is, higher education enrollment rate is high, but university graduate employment rate is low. To eliminate this cycle and relieve youth unemployment and young people to enter the labor market early, dual education and training system is needed. This dual system can support working and learning in parallel. So, worker can get the opportunity pre-employment and post-learning and improve his/her job skills. Recent MOOC (Massive Open On-line Course), a new form of online education system, has emerged. MOOC combines education, entertainment and social networking, and emphasize the interaction between faculty and student and between students. The educational contents of MOOC are available free of charge. Using newly changed online education environments we can effectively provide knowledge and skills. In technology and engineering education hands-on training is necessary. In order to support work based learning dual system for worker to work and learn in parallel, we should build the multi-learning system to combine the online education and campus hands-on practice.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.1
• /
• pp.219-223
• /
• 2020

To guarantee AI model's prominent recognition rate and recognition precision, obtaining the large number of data is essential. In this paper, we propose transfer learning-based object detection algorithm for maintaining outstanding performance even when the volume of training data is small. Also, we proposed a tranfer learning network combining Resnet-50 and YOLO(You Only Look Once) network. The transfer learning network uses the Leeds Sports Pose dataset to train the network that detects the person who occupies the largest part of each images. Simulation results yield to detection rate as 84% and detection precision as 97%.

• Journal of Gastric Cancer
• /
• v.15 no.3
• /
• pp.151-158
• /
• 2015

Purpose: Obesity is associated with morbidity following gastric cancer surgery, but whether obesity influences morbidity after laparoscopic gastrectomy (LG) remains controversial. The present study evaluated whether body mass index (BMI) and visceral fat area (VFA) predict postoperative complications. Materials and Methods: A total of 217 consecutive patients who had undergone LG for gastric cancer between May 2003 and December 2005 were included in the present study. We divided the patients into two groups ('before learning curve' and 'after learning curve') based on the learning curve effect of the surgeon. Each of these groups was sub-classified according to BMI (<$25kg/m^2$ and ${\geq}25kg/m^2$) and VFA (<$100cm^2$ and ${\geq}100cm^2$). Surgical outcomes, including operative time, quantity of blood loss, and postoperative complications, were compared between BMI and VFA subgroups. Results: The mean operative time, length of hospital stay, and complication rate were significantly higher in the before learning curve group than in the after learning curve group. In the subgroup analysis, complication rate and length of hospital stay did not differ according to BMI or VFA; however, for the before learning curve group, mean operative time and blood loss were significantly higher in the high VFA subgroup than in the low VFA subgroup (P=0.047 and P=0.028, respectively). Conclusions: VFA may be a better predictive marker than BMI for selecting candidates for LG, which may help to get a better surgical outcome for inexperienced surgeons.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.5
• /
• pp.619-625
• /
• 1995

This paper constructs a feed-forward learning complex algorithm which replaced by the backpropagation learning. This algorithm first attempts to organize the pattern vectors into clusters by Generalized Learning Vector Quantization(GLVQ) clustering algorithm(Nikhil R. Pal et al, 1993), second, regroup the pattern vectors belonging to different clusters, and the last, recognize into regrouping pattern vectors by single layer perceptron. Because this algorithm is feed-forward learning algorithm, time is less than backpropagation algorithm and the recognition rate is increased. We use 250 ASCII code bit patterns that is normalized to 16$\times$8. As experimental results, when 250 patterns devide by 10 clusters, average iteration of each cluster is 94.7, and recognition rate is 100%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.5
• /
• pp.591-599
• /
• 2020

Deep learning is widely utilized to classify or recognize objects of real-world. An abundance of data is trained on high-performance computers and a trained model is generated, and then the model is loaded in an inferencer. The inferencer is used in various environments, so that it may cause unrecognized objects or low-accuracy objects. To solve this problem, real-world objects are collected and they are trained periodically. However, not only is it difficult to immediately improve the recognition rate, but is not easy to learn an inferencer on embedded devices. We propose a real-time handwriting recognizer based on partial learning on embedded devices. The recognizer provides a training environment which partially learn on embedded devices at every user request, and its trained model is updated in real time. As this can improve intelligence of the recognizer automatically, recognition rate of unrecognized handwriting increases. We experimentally prove that learning and reasoning are possible for 22 numbers and letters on RK3399 devices.