• Journal of the Korean Society of Radiology
• /
• v.18 no.1
• /
• pp.21-26
• /
• 2024

The one of the famous deep learning models for object detection task is you only look once version 5 (YOLOv5) framework based on the one stage architecture. In addition, YOLOv5 model indicated high performance for accurate lesion detection using the bottleneck CSP layer and skip connection function. The purpose of this study was to evaluate the performance of YOLOv5 framework according to various hyperparameters in position emission tomogrpahy (PET) phantom images. The dataset was obtained from QIN PET segmentation challenge in 500 slices. We set the bounding box to generate ground truth dataset using labelImg software. The hyperparameters for network train were applied by changing optimization function (SDG, Adam, and AdamW), activation function (SiLU, LeakyRelu, Mish, and Hardwish), and YOLOv5 model size (nano, small, large, and xlarge). The intersection over union (IOU) method was used for performance evaluation. As a results, the condition of outstanding performance is to apply AdamW, Hardwish, and nano size for optimization function, activation function and model version, respectively. In conclusion, we confirmed the usefulness of YOLOv5 network for object detection performance in nuclear medicine images.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.11a
• /
• pp.62-65
• /
• 2011

최근 화두가 되고 있는 스마트 폰 앱의 관심으로 스마트 TV의 앱에 대한 관심도 함께 증가하고 있다. TV시청 이용자들의 편의를 위해 증가하고 있는 수많은 채널과 콘텐츠 중, 개인 사용자의 이용 습관 및 대중의 선호 프로그램을 고려하여, 편리하게 원하는 TV프로그램에 접근하도록 해 주는 TV 앱이 있다면 이는 매우 중요한 기능으로 자리 잡을 가능성이 높을 것으로 예상된다. 이에 본 논문은 사용자의 시청 이용행태를 기반으로 주제모델링 기술의 고전적 모델인 LDA을 기반으로 협업필터링을 결합한 TV 선호 프로그램 추천 알고리듬을 제안한다. 개인의 관심 선호도는 일반적으로 특정 개수로 한정지어지는 특성을 고려하여, 개인 선호도 특성이 구별 되도록 두 가지 방법을 적용하였다. 하나는 개인 선호도 프로파일의 특정 상위 주제만을 고려하는 것이고, 또 다른 하나는 개인별 주제에 대한 선호도의 다양성이 드러나도록 비대칭 하이퍼-파라미터를 갖는 LDA를 사용 하였다. 실험 결과, 두 가지 방식에 대해 사용자의 실제 TV시청 이용내역 데이터를 기반으로 추천 성능의 향상을 평균 Precision 값을 측정하여 확인하였다. 또한, 본 논문에서는 주제 모델링을 통해 학습된 각 주제의 상위 확률의 TV 프로그램들을 분석한 결과, 하나의 주제가 개인별 시청의 특성 보다는 가족단위의 시청 특성을 드러냄을 확인할 수 있었다.

• Journal of Korean Society of Water and Wastewater
• /
• v.35 no.6
• /
• pp.507-516
• /
• 2021

In this study, the machine learning which has been widely used in prediction algorithms recently was used. the research point was the CD(chudong) point which was a representative point of Daecheong Lake. Chlorophyll-a(Chl-a) concentration was used as a target variable for algae prediction. to predict the Chl-a concentration, a data set of water quality and quantity factors was consisted. we performed algorithms about random forest and gradient boosting with Python. to perform the algorithms, at first the correlation analysis between Chl-a and water quality and quantity data was studied. we extracted ten factors of high importance for water quality and quantity data. as a result of the algorithm performance index, the gradient boosting showed that RMSE was 2.72 mg/m³ and MSE was 7.40 mg/m³ and R² was 0.66. as a result of the residual analysis, the analysis result of gradient boosting was excellent. as a result of the algorithm execution, the gradient boosting algorithm was excellent. the gradient boosting algorithm was also excellent with 2.44 mg/m³ of RMSE in the machine learning hyperparameter adjustment result.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.4
• /
• pp.111-117
• /
• 2022

Many studies have been conducted on software fault prediction models for decades, and the models using machine learning techniques showed the best performance. Deep learning techniques have become the most popular in the field of machine learning, but few studies have used them as classifiers for fault prediction models. Some studies have used deep learning to obtain semantic information from the model input source code or syntactic data. In this paper, we produced several models by changing the model structure and hyperparameters using MLP with three or more hidden layers. As a result of the model evaluation experiment, the MLP-based deep learning models showed similar performance to the existing models in terms of Accuracy, but significantly better in AUC. It also outperformed another deep learning model, the CNN model.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.401-403
• /
• 2021

본 연구는 한국지능정보사회진흥원에서 제공한 2018년 스마트폰 과의존 실태조사에서 사용된 11개 변수와 스마트폰 과의존과의 관계를 탐색하고, 이를 통해 딥러닝 기반 스마트폰 과의존 분류 분석 모델을 개발하고자 시행되었다. 학습데이터셋은 전국 10,000개 가구내 만 3-69세 스마트폰 이용자 25,465명의 스마트폰 이용 형태 및 개인적 특성에 관한 데이터이다. 딥러닝은 심층신경망(DNN)을 설계하였으며, 은닉층(hidden layer)은 4개층으로 구성하였다. 입력한 데이터는 각각 200개, 150개, 100개, 50개, 2개 노드를 거치면서 최종 출력 정보인 스마트폰 과의존 분류율로 나타나는 모델이다. 이때 스마트폰 과의존 분류률을 높이기 위해 학습률(learning rate)과 같은 하이퍼 파라미터를 활용하여 세부조정하면서 가장 잘 학습하는 값을 찾아내었다. 연구결과, 학습횟수가 300번으로 학습율(learning.rate)이 0.01일때 훈련데이터에서 97.43%, 검증데이터에서 98.06%로 가장 높게 나타났다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.394-396
• /
• 2021

Additional pathological tests using imaging equipment are essential before diagnosing cancer cells. Recently, in order to reduce the need for time and human resources in these fields, research related to the establishment of a system capable of automatic classification of cancer cells using artificial intelligence is being actively conducted. However, in both previous studies, there were relatively limited deep learning algorithms and cell types, and limitations existed with low accuracy at the same time. In this study, a method of performing 4class Classification on four types of cancer cells through the Convolution Neral Network, a type of in-depth learning. EfficientNet, ResNet, and Inception were used, and finally Resnet was used to obtain an accuracy of 96.11 on average for k-fold.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.5
• /
• pp.949-958
• /
• 2020

The functions of shock absorbers are to dampen body, suspend motions, dissipate impact energy, and control tire force variation. During the operation, hydraulic oil is passed between the chambers via a flow restrictions. Therefore the damping force characteristics of shock absorber is determined by the characteristics of orifices and flow restrictions. The uncertainty in design variable affects the performance of suspension system strongly. But, the researches about the influence of uncertainty in design variable such as a fluid restriction's property of shock absorber, on the suspension system performance was hardly ever proposed. In this paper, we used statistical method of Latin Hypercube sampling, and the effects of design variables uncertainty on the performance of suspension system was presented.

• Journal of Advanced Navigation Technology
• /
• v.28 no.1
• /
• pp.149-154
• /
• 2024

Research on particulate matter is advancing in real-time, and various methods are being studied to improve the accuracy of prediction models. Furthermore, studies that take into account various factors to understand the precise causes and impacts of particulate matter are actively being pursued. This paper trains an LSTM model using seasonal data and another LSTM model using concentration-based data. It compares and analyzes the PM_2.5 prediction performance of the two models. To train the model, weather data and air pollutant data were collected. The collected data was then used to confirm the correlation with PM_2.5. Based on the results of the correlation analysis, the data was structured for training and evaluation. The seasonal prediction model and the concentration-specific prediction model were designed using the LSTM algorithm. The performance of the prediction model was evaluated using accuracy, RMSE, and MAPE. As a result of the performance evaluation, the prediction model learned by concentration had an accuracy of 91.02% in the "bad" range of AQI. And overall, it performed better than the prediction model trained by season.

• Korean Journal of Construction Engineering and Management
• /
• v.21 no.6
• /
• pp.38-45
• /
• 2020

This study presents a systematic procedure for developing a short-term prediction deep learning model of rebar price using bidirectional LSTM, Random Search, data combination, Dropout. In general, users intuitively determine these values, making it time-consuming and repetitive attempts to explore results with good predictive performance, and the results found by these attempts cannot be guaranteed to be excellent. With the proposed approach presented in this study, the average accuracy of short-term price forecasts is approximately 98.32%. In addition, this approach could be used as basic data to produce good predictive results in a study that predicts prices with time series data based on statistics, including building materials other than rebars.

• The Journal of the Korea Contents Association
• /
• v.20 no.10
• /
• pp.702-714
• /
• 2020

Stock investment is a personal investment technique that has gathered tremendous interest since the reduction in interest rates and tax exemption. However, it is risky especially for those who do not have expert knowledge on stock volatility. Therefore, it is well understood that accurate stock trend prediction can greatly help stock investment, giving birth to a volume of research work in the field. In order to compare different research works and to optimize hyper-parameters for prediction models, it is required to have an evaluation standard that can accurately assess performances of prediction models. However, little research has been done in the area, and conventionally used methods have been employed repeatedly without being rigorously validated. For this reason, we first analyze performance evaluation of stock trend prediction with respect to performance metrics and data composition, and propose a fair evaluation method based on prediction disparity ratio.