• Journal of Korea Multimedia Society
• /
• v.25 no.2
• /
• pp.157-165
• /
• 2022

As the stock market is affected by various circumstances including economic and political variables, predicting the stock market is considered a still open problem. When combined with corporate financial statement data analysis, which is used as fundamental analysis, and technical analysis with a short data generation cycle, there is a problem that the time domain does not match. Our proposed method, LSTE the operating profit and market outlook of a petrochemical company and estimates the sales and operating profit of the company, it was possible to solve the above-mentioned problems and improve the accuracy of stock price prediction. Extensive experiments on real-world stock data show that our method outperforms the 8.58% relative improvements on average w.r.t. accuracy.

• Annual Conference of KIPS
• /
• 2023.11a
• /
• pp.691-694
• /
• 2023

본 논문은 CCTV 화면에서의 다양한 이상상황 중 교량 데이터에 특화된 자세 추정 기반 이상탐지 알고리즘을 소개한다. 교량은 크게 도로, 인도 이렇게 두 구역으로 나눠지며, 사람들의 이동방향이 한정적이라는 특징을 가지는 장소 중 하나이다. 이러한 장소적 특징을 이용하고자 사람 자세 추정을 통해 이상의 기준을 잡고 교량 데이터에 특화된 이상탐지 알고리즘을 제안한다. CCTV 영상은 이상을 정하기 어렵고 이상에 대한 레이블이 없는 데이터가 대부분이며 이상에 대한 레이블 생성시 많은 비용 발생이 필수적이다. 본 연구에서는 이러한 한계점을 극복하고자 영상 데이터를 이미지 단위가 아닌 영상 단위로 레이블이 담긴 weakly label 을 가지는 데이터를 활용한 이상탐지 모델을 이용하였다. 특히, 교량에서의 이상상황의 특징인 사람 자세 추정으로 추출한 특질을 추가하여 기존 알고리즘의 이상탐지 예측 성능을 개선하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.99-99
• /
• 2011

최근에 미국에서 출범한 수질 교환 법은 수질기준을 혁신적인 접근방법으로 만족시키는 법이다. 이 법안은 수질기준을 초과하지 않는 조건에서 한 유역 안 다른 지점들의 점 오염과 비점오염 배출의 교환을 허용한다. 이 법안을 적용하기 위한 방법을 도출하기 위하여 많은 시험 프로그램을 운영하고 있지만 여전히 실제 교환은 상대적으로 적게 이루어지고 있다. 또한 하천내의 비 점오염량의 불확실성으로 인하여 교환 지점을 선정하고 적용하는 데에 큰 어려움이 있다. Hydrological Simulation Program-Fortran (HSPF)은 Soil and Water Assessement Tool (SWAT)과 함께 미국 하천 모델링에 많이 쓰이는 유역 모델로써 특히 HSPF는 각각의 토지 피복도의 퍼센트 투수량을 지정함으로써 도시지역의 유출량을 시뮬레이션 하는데 강점이 있다. 미국 중서부 미조리 주의 퍼시픽시를 포함하고 있는 Brush Creek 유역을 선택하여 퍼시픽시의 도시화 증가로 인한 Brush Creek 유역의 상류와 하류지역의 유출량 및 Sediment 변화를 예측하여 수질관리법을 적용하는 방법을 연구하였다. 이 연구의 특징은 원격탐사 이미지 (QuickBird)로 구현한 최근의 토지 이용을 미래의 도시지역으로 전환한 토지이용도를 사용함으로서 특정 유역을 가장 정확하게 이해하는 시뮬레이션을 가능하도록 한다는 점이다. 각각의 토지이용에서 도시화가 3가지의 강도를 가지고 진행된다는 시나리오를 이용하여 모델링을 하였고 이로 인해 계산된 유출량과 Sediment 양을 이용하여 각각의 토지이용 변화 별 수질 교환단위를 도출 하였다.

• Journal of Platform Technology
• /
• v.10 no.1
• /
• pp.20-28
• /
• 2022

Artificial intelligence-based health data verification has become an essential element not only to help clinical research, but also to develop new treatments. Since the US Food and Drug Administration (FDA) approved the marketing of medical devices that detect mild abnormal diabetic retinopathy in adult diabetic patients using artificial intelligence in the field of medical diagnosis, tests using artificial intelligence have been increasing. In this study, an artificial intelligence model based on image classification was created using a Teachable Machine supported by Google, and a predictive model was completed through learning. This not only facilitates the early detection of cataracts among eye diseases occurring among patients with chronic diseases, but also serves as basic research for developing a digital personal health healthcare app for eye disease prevention as a healthcare program for eye health.

• Journal of Aerospace System Engineering
• /
• v.16 no.4
• /
• pp.17-27
• /
• 2022

In this study, a deep learning-based network that can predict the aerodynamic characteristics of airfoils was designed, and the feasibility of the proposed network was confirmed by applying aerodynamic data generated by Xfoil. The prediction of aerodynamic characteristics according to the variation of airfoil thickness was performed. Considering the angle of attack, the coordinate data of an airfoil is converted into image data using signed distance function. Additionally, the distribution of the pressure coefficient on airfoil is expressed as reduced data via proper orthogonal decomposition, and it was used as the output of the proposed network. The test data were constructed to evaluate the interpolation and extrapolation performance of the proposed network. As a result, the coefficients of determination of the lift coefficient and moment coefficient were confirmed, and it was found that the proposed network shows benign performance for the interpolation test data, when compared to that of the extrapolation test data.

• Spatial Information Research
• /
• v.22 no.4
• /
• pp.39-47
• /
• 2014

The Yeongsan river estuary has a serious water quality problem due to the water stagnation and it is imperative to predict the changes of water quality for mitigating water pollution. EFDC(Environmental Fluid Dynamics Code) model was mainly utilized to predict the changes of water quality for the estuary. The EFDC modeling normally accompanies the large volume of modeling output. For checking the spatial distribution of the modeling results, post-processing for converting of the output is prerequisite and mainly post-processing program is EFDC_Explorer. However, EFDC_Explorer only shows the spatial distribution of the time series and this doesn't support overlay function with other thematic maps. This means the impossible to the connection analysis with a various GIS data and high dimensional analysis. Therefore, this study aims to develop a post-processing system of a EFDC output to use them as GIS layers. For achieving this purpose, a editing module for main input files, and a module for converting binary format into an ASCII format, and a module for converting it into a layer format to use in a GIS based environment, and a module for visualizing the reconfigured model result efficiently were developed. Using the developed system, result file is possible to automatically convert the GIS based layer and it is possible to utilize for water quality management.

• Journal of Internet Computing and Services
• /
• v.22 no.2
• /
• pp.29-39
• /
• 2021

As damages to individuals, private sectors, and businesses increase due to newly occurring cyber attacks, the underlying network security problem has emerged as a major problem in computer systems. Therefore, NIDS using machine learning and deep learning is being studied to improve the limitations that occur in the existing Network Intrusion Detection System. In this study, a deep learning-based NIDS model study is conducted using the Convolution Neural Network (CNN) algorithm. For the image classification-based CNN algorithm learning, a discrete algorithm for continuity variables was added in the preprocessing stage used previously, and the predicted variables were expressed in a linear relationship and converted into easy-to-interpret data. Finally, the network packet processed through the above process is mapped to a square matrix structure and converted into a pixel image. For the performance evaluation of the proposed model, NSL-KDD, a representative network packet data, was used, and accuracy, precision, recall, and f1-score were used as performance indicators. As a result of the experiment, the proposed model showed the highest performance with an accuracy of 85%, and the harmonic mean (F1-Score) of the R2L class with a small number of training samples was 71%, showing very good performance compared to other models.

• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.31-43
• /
• 2022

Traditional video compression has developed so far based on hybrid compression methods through motion prediction, residual coding, and quantization. With the rapid development of technology through artificial neural networks in recent years, research on image compression and video compression based on artificial neural networks is also progressing rapidly, showing competitiveness compared to the performance of traditional video compression codecs. In this paper, a new method capable of improving the performance of such an artificial neural network-based video compression model is presented. Basically, we take the rate-distortion optimization method using the auto-encoder and entropy model adopted by the existing learned video compression model and shifts some components of the latent information that are difficult for entropy model to estimate when transmitting compressed latent representation to the decoder side from the encoder side, and finally compensates the distortion of lost information. In this way, the existing neural network based video compression framework, MFVC (Motion Free Video Compression) is improved and the BDBR (Bjøntegaard Delta-Rate) calculated based on H.264 is nearly twice the amount of bits (-27%) of MFVC (-14%). The proposed method has the advantage of being widely applicable to neural network based image or video compression technologies, not only to MFVC, but also to models using latent information and entropy model.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.39 no.7
• /
• pp.557-566
• /
• 2015

The characteristics of a gas-liquid Taylor (slug) flow in a square micro-channel with dimensions of $600{\mu}m{\times}600{\mu}m$ are experimentally investigated in this paper. The test fluids were nitrogen and water. The superficial velocities of the liquid and gas were in the ranges of 0.01 - 3 m/s and 0.1 - 3 m/s, respectively. The bubble and liquid slug lengths, bubble velocities, and bubble frequencies for various inlet conditions were measured by analyzing optical images obtained with a high-speed camera. It was found that the measured values (bubble and liquid slug lengths, bubble velocities) were not in good agreement with the values obtained using empirical models presented in the existing literature. Modified models for the bubble and liquid slug lengths and bubble velocity are suggested and shown to be in good agreement (${\pm}20$) with the measured values. Moreover, the bubble frequency could be predicted well by the relationship between the unit cell length and its velocity.

• Archives of design research
• /
• v.17 no.1
• /
• pp.253-264
• /
• 2004

The final goal of this research is development of graphic design evaluation methodology in elevation of a usability at the mobile internet services and of measurement model which can forecast user needs in interface design, and systemize evaluation basis. For this, we systemize core contents of GUI design evaluation methodology and embodied UI design support system that supports prototype layout and evaluation process directly. The sight language that can inform flow of controled information by the quick and implicated method so that user may complete task in a short time without overload of recognition in limited display environment of Small Screen device it must improve objectivity in the reflection of UI design with image. Thus evaluation methodology that can evaluate usability of mobile internet systematically is important and specially, graphic design evaluation model which can forecast user's design need and trend is meaningful because of special quality that can reflect sensitive aspect of user in interface design. Mobile internet GUI was done by the result of this design evaluation, and I hope this result can be utilized for the GUI development of Ubiquitous environment for the future research.