• Journal of the Korea Society of Computer and Information
• /
• v.8 no.3
• /
• pp.113-123
• /
• 2003

Many researches on the application of neural networks for making financial index prediction have proven their advantages over statistical and other methods. In this paper, a neural network model is proposed for the Buying, Holding or Selling timing prediction in stocks by the price index of stocks by inputting the closing price and volume of dealing in stocks and the technical indexes(MACD, Psychological Line). This model has an equalized multi-layer arithmetic function as well as the time series prediction function of backpropagation neural network algorithm. In the case that the numbers of learning data are unbalanced among the three categories (Buying, Holding or Selling), the neural network with conventional method has the problem that it tries to improve only the prediction accuracy of the most dominant category. Therefore, this paper, after describing the structure, working and learning algorithm of the neural network, shows the equalized multi-layer arithmetic method controlling the numbers of learning data by using information about the importance of each category for improving prediction accuracy of other category. Experimental results show that the financial index prediction using the equalized multi-layer arithmetic neural network has much higher correctness rate than the other conventional models.

• Journal of KIISE:Databases
• /
• v.29 no.5
• /
• pp.381-392
• /
• 2002

This paper deals with the problem of shape-based retrieval in time-series databases. The shape-based retrieval is defined as the operation that searches for the (sub)sequences whose shapes are similar to that of a given query sequence regardless of their actual element values. In this paper, we propose an effective and efficient approach for shape-based retrieval of subsequences. We first introduce a new similarity model for shape-based retrieval that supports various combinations of transformations such as shifting, scaling, moving average, and time warping. For efficient processing of the shape-based retrieval based on the similarity model, we also propose the indexing and query processing methods. To verify the superiority of our approach, we perform extensive experiments with the real-world S&P 500 stock data. The results reveal that our approach successfully finds all the subsequences that have the shapes similar to that of the query sequence, and also achieves significant speedup up to around 66 times compared with the sequential scan method.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.290-290
• /
• 2019

최근 우리나라 수자원 오염의 심각성에 관한 인식이 커지고 있으며, 수질 및 조류 관리의 필요성이 다양한 분야에서 증가하고 있다. 우리나라의 경우 초여름부터 장마 이전과 장마 이후부터 10월까지 호소의 영양 단계가 조류의 성장에 충분한 상태이기 때문에 남조류가 대량으로 증식된다. 하지만 기상 조건에 따라 조류의 발생 시기와 발생 정도는 매년 다르게 나타날 수 있다. 한편, 조류를 포함한 수질 모의 모형은 현재의 상태에 대한 정확한 모사를 통해 수질 관리와 미래의 예측 등 다양한 방면으로 활용된다. 따라서 복잡한 현상을 모형을 통해 재현하는 것은 중요하다. 본 연구에서는 대청호 지역을 대상으로 CE-QUAL-W2 모형을 이용하여 2016년 1월부터 2018년 7월까지 수질 및 조류의 거동을 모의하였다. 금강의 지류 하천인 소옥천에 위치한 옥천천 측정소부터 대청호 내부의 대청댐5 구간을 상류 및 하류 경계로 설정하였고, 수질 및 조류의 모의를 위해 물환경 정보시스템과 실시간 수질 정보시스템의 관측자료 및 대청댐의 운영 자료, 기상청의 기상관측 자료를 수집하였다. 모형의 안정화(warm up) 기간을 고려하여 2015년 1월부터 2018년 7월까지 주요 수질 항목인 수온, 용존산소, 총 인, 총 질소, 클로로필-a에 대한 분석을 수행하였고, 연구대상 지역 내의 대청댐6 지점의 관측값과 모의 결과를 비교 분석하였다. 수온과 용존산소, 클로로필-a의 경우 모의 결과가 관측 값을 비교적 잘 모사하였지만, 총 인과 총 질소의 경우 자동 수질 측정지점의 값과 일반수질 측정 지점의 측정 방법의 차이로 인해 오차가 다소 발생하는 것을 확인할 수 있었다. 본 연구는 CE-QUAL-W2 모형을 이용한 수질 및 조류 모의에 관한 연구로서, CE-QUAL-W2 모형을 이용한 수질 및 조류 관리에 활용될 수 있을 것으로 판단된다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.101-101
• /
• 2016

하천원격탐사는 원격탐사의 하위 개념으로서 계측하고자 하는 대상인 하천이나 호소 수체에서 발생하는 빛의 반사, 복사 또는 방출되는 양을 획득하고 분석하여 수리량, 지형 등 하천 조사에 활용하는 기법이다. 일반적으로 원격탐사는 주로 위성영상 자료를 활용하여 수행되어 자료취득비용이 고가이고 해외 위성자료에 의존하여 시공간적인 해상도가 매우 낮아 유역에 비해 공간적인 규모가 작고 변동 시간이 짧은 하천에 적용하는 데 한계가 있어 왔다. 또한, 단순한 사진촬영으로 도출할 수 있는 정보에 한계가 있고 자료를 저장 및 분석할 수 있는 기법도 부족하여 하천조사에 원격탐사를 활용한 사례가 드물었다. 그러나, 최근 드론과 같은 운반체 기술이 획기적으로 개선되고 있고 다양한 영상촬영장비의 개발과 IT기술의 발전으로 인해 위성영상에 비해 시공간적 해상도가 매우 정밀한 자료를 저렴한 비용으로 획득 가능해졌다. 또한, 매우 조밀한 파장대로 세분된 빛의 세기를 측정할 수 있는 초분광 영상을 이용한 원격탐사기법도 하천과 같은 좁은 영역에 적용이 가능해졌다. 초분광영상은 가시광선 외에 자외선과 적외선 영역에 해당하는 반사광을 200개 이상의 조밀한 파장대로 나누어 측정할 수 있어 수리량, 하상, 식생 등 하천 수체와 관련된 정보를 조사할 가능성이 증가하였다. 본 연구에서는 하천 수체에서 취득한 초분광 영상을 이용하여 하천특성과의 상관관계를 규명하고 이를 통해 초분광 영상 기반의 하천특성 계측 기법을 개발하고자 하였다. 드론과 같은 항공영상에 적용하기 전에, 우선 지상에서 측정된 초분광 영상과 하상재료 및 수심과의 상관관계를 규명하여 초분광 영상의 하천조사로의 사용 가능성을 점검해 보았다. 폭 10m, 수심 1m의 저수심의 소하천에 적용한 결과, 초분광 영상의 표준화 및 패턴 분석을 통해 수중에 위치한 하상재료를 구분할 수 있었고 주성분분석 등을 통해 수심과 상관성도 일부 도출되어 하천조사에 초분광영상이 활용될 수 있음을 확인하였다.

• Journal of Broadcast Engineering
• /
• v.26 no.4
• /
• pp.390-397
• /
• 2021

As digital twin technology is applied to various industrial fields, technologies to effectively process large amounts of data are required. In this paper, we discuss a customized service method for fast search and effective delivery of large-scale data for underground facility for public utilities management. The proposed schemes are divided into two ways: a fast search data generation method and a customized information service segmentation method to efficiently search and abbreviate vast amounts of data. In the high-speed search data generation, we discuss the configuration of the synchronization process for the time series analysis of the sensors collected in the underground facility and the additional information method according to the data reduction. In the user-customized service method, we define the types of users in normal and disaster situations, and discuss how to service them accordingly. Through this study, it is expected to be able to develop a systematic data generation and service model for the management of underground utilities that can effectively search and receive large-scale data in a disaster situation.

• Journal of the Society of Disaster Information
• /
• v.16 no.4
• /
• pp.842-848
• /
• 2020

Purpose: It is important to operate safely and economically in obsolescent power plant facilities. Economical operation is related in the balance of the supply and demand. Safety operation predicts the possible risks in the facilities and then, takes measures to the facilities. For the monitoring of the power plant facilities, we needs several kinds of the sensing system. From the sensors data, we can predict the possible risk. Method: We installed the acoustic, vibration, electric and smoke sensors in the power plant facilities. Using the data, we developed 3 kinds of prediction models, such as, demand prediction, plant engine abnormal prediction model, and risk prediction model. Results: Accuracy of the demand prediction model is over 90%. The other models make a stable operation of the system. Conclusion: For the sustainable operation of the obsolescent power plant, we developed 3 kinds of AI prediction models. The model apply to JB company's power plant facilities.

• Journal of Convergence for Information Technology
• /
• v.11 no.9
• /
• pp.84-93
• /
• 2021

The purpose of this study is to analyze whether the spread of COVID-19 infectious diseases acts as a fear to investors and affects the direction and volatility of stock returns. The investor fear index was proposed using the domestic confirmed patient information of COVID-19, and the influence on stock prices was empirically analyzed. The direction and volatility models of stock prices used the Granger causality and GARCH models, respectively. The results of empirical analysis using the KOSPI index from February 20, 2020 to June 30, 2021 are as follows: First, the COVID-19 fear index showed causality to future stock prices. Second, the COVID-19 fear index has a negative effect on the volatility of KOSPI index returns. In future studies, it is necessary to document the cause by using individual business performance and stock price instead of the stock index.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.4
• /
• pp.503-509
• /
• 2022

The purpose of this study is to set the rate of change between the market price of the next day and the previous day to be predicted as the predicted value, and the market price for each section is generated by dividing the stock price ranking of the next day to be predicted at regular intervals, which is different from the previous papers that predict the market price. We would like to propose a new time series data prediction method that predicts the market price change rate of the final next day through a model using the rate of change as the predicted value. The change in the performance of the model according to the degree of subdivision of the predicted value and the type of input data was analyzed.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.5
• /
• pp.869-879
• /
• 2022

Through this study, we discovered that among three types of compressed data blocks generated through the Deflate algorithm, No-Payload Non-Compressed Block type (NPNCB) which has no literal data can be randomly generated and inserted between normal compressed blocks. In the header of the non-compressed block, there is a data area that exists only for byte alignment, and we called this area as DBA (Disposed Bit Area), where an attacker can hide various malicious codes and data. Finally we found the vulnerability that hides malicious codes or arbitrary data through inserting NPNCBs with infected DBA between normal compressed blocks according to a pre-designed attack scenario. Experiments show that even though contaminated NPNCB blocks were inserted between normal compressed blocks, commercial programs decoded normally contaminated zip file without any warning, and malicious code could be executed by the malicious decoder.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.6
• /
• pp.1181-1191
• /
• 2021

The pandemic has rapidly developed a non-face-to-face environment. However, the sudden transition to a non-face-to-face environment has led to new security issues in various areas. One of the new security issues is the security threat of insiders, and the zero trust security model is drawing attention again as a technology to defend against it.. Software Defined Perimeter (SDP) technology consists of various security factors, of which device validation is a technology that can realize zerotrust by monitoring insider usage behavior. But the current SDP specification does not provide a technology that can perform device validation.. Therefore, this paper proposes a device validation technology using SVDD-based abnormal behavior detection technology through user behavior monitoring in an SDP environment and presents a way to perform the device validation technology in the SDP environment by conducting performance evaluation.