• 한국컴퓨터정보학회논문지
• /
• 제25권5호
• /
• pp.47-56
• /
• 2020

주가지수는 한 국가의 경제 지표뿐만 아니라 투자판단의 지표로도 활용되므로 이를 예측하는 연구가 지속해서 진행되고 있다. 주가지수 예측을 하는 작업은 기술적, 경제적 및 심리적 요인 등이 반영된 것으로 예측의 정확도를 위해서는 복합적 요인을 고려해야 한다. 따라서 지수의 변동에 영향을 미치는 요인들을 선별하여 반영한 주가지수 예측모델연구가 필요하다. 이와 관련한 기존 연구에서는 시장의 변동을 만들어 내는 뉴스 정보 또는 거시 경제 지표를 각각 이용하거나, 몇 가지의 지표 조합만을 반영한 예측 연구가 대부분이었다. 따라서 본 연구에서는 미국 다우존스지수 예측을 위해 뉴스 정보의 감성 분석과 다양한 거시경제지표를 고려하여 효과적인 지표 조합을 제시하고자 한다. 뉴스 정보의 감성 분석은 최신 자연어처리 기법인 BERT와 NLTK VADER를 사용하고, 예측모델은 주가예측모델로 적합하다고 알려진 딥러닝 예측모델 LSTM을 적용하여 가장 효과적인 지표 조합을 제시했다.

• 자원ㆍ환경경제연구
• /
• 제19권4호
• /
• pp.733-752
• /
• 2010

본고는 한국의 화력발전소의 생산성 변화율 자체와 그 변화행태와 발전유형별 분포, 그리고 생산성 변화의 구성요소 등을 $CO_2$ 배출량을 감안하여 분석하며, 그러한 분석목적을 달성하는 데 매우 유용한 Luenberger지수를 비모수적 기법을 이용해 도출하는 방식으로 연구를 진행한다. 분석결과 $CO_2$ 배출량을 감안하지 않을 때에는 생산성 변화율이 상당한 정도 왜곡된 형태로 나타날 수 있음이 확인되었다. 특히 발전소의 경우 가동률 차이가 생산성 지표에 큰 영향을 미치게 되는데, $CO_2$ 배출량을 감안하지 않으면 주로 가동률 차이에 의한 생산량 변화가 생산성 변화율에 영향을 미치게 되고, 그 결과 생산성 변화율 계측치가 비상식적으로 높거나 낮게 되는 현상이 발견되었다. 아울러 분석기간 동안 $CO_2$의 배출량이 늘어났기 때문에 이를 반영하지 않고 생산성 증가율을 계측하면 생산성 증가율을 과대평가하게 됨도 밝혀졌다. $CO_2$를 모형에 포함하느냐의 여부는 또한 발전소 유형별로 생산성 증가율이 어떻게 다른지, 그리고 생산성 변화가 발생한 주요인은 무엇인지를 설명함에 있어서도 상당한 차이를 초래하였기 때문에 이들 내용을 확인하기 위해서도 적절한 방식으로 $CO_2$ 배출량을 반영해 주는 것이 중요하다.

• 기술혁신연구
• /
• 제25권4호
• /
• pp.1-15
• /
• 2017

2014년, 금융위원회는 기술신용평가기관(TCB, Tech Credit Bureaus)을 지정하여 기술신용평가서를 발급하게 하였고 현재까지 5개의 기술신용평가기관과 금융위원회 권고, 레벨 4에 진입한 KEB하나은행, 국민은행, 우리은행, 신한은행 등에서 기술신용평가서를 발급하고 있다. 한편, KEB하나은행의 기술평가모델은 25개의 세부평가항목으로 구성되어 있으며, 이러한 항목등급이 가중 결합되어 기술등급이 산출, 기술등급은 신용등급과 결합하여 최종적으로 기술신용등급이 산출된다. 본 연구에서는 KEB하나은행에서 2016년 하반기에 자체발급한 406건의 기술평가결과를 분석하였으며, 경영주 동업종 근무년수, 기술개발전담부서 보유여부, 기술인력, 연구개발투자금액, 인증수, 특허수를 기반으로 지표간의 상관분석 및 기술등급과의 영향력을 분석하였다. 분석결과에 의하면, 기술개발전담부서, 특허수, 연구개발투자금액 등의 정량적지표가 기업 기술등급에 상당한 영향을 끼치는 것으로 나타났으며, 특히, 기술개발전담부서 보유여부는 기술등급과 가장 높은 상관관계를 나타내고 있음을 나타냈다.

• 펄프종이기술
• /
• 제48권1호
• /
• pp.93-99
• /
• 2016

The extraction efficiency depending on the extracting methods and the conditions of extraction was investigated. The common steam extraction was compared to the distillation extraction method. The effects of the samples size and the extraction time on the extract yield were also investigated by using UV-Vis spectrophotometer. One of the functional components of pine needle extract as the natural phenol base components were detected by the UV-VIS at around 235 nm wavelength range. The absorbance intensity at around 235 nm wavelength of the pine needle extract was used as the indicator of the extraction efficiency in this experiment. The distillation extraction showed the higher extract yield than the steam extraction. The grinding treatment of pine needles also resulted in the better extract performance, but the severe grinding showed a little decrease in the extract yield especially in case of the distillation extraction method. More than half of the extract was collected at the first stage of the extraction, that was the first 15 minutes in the total 60 minutes extraction.

• 대한산업공학회지
• /
• 제37권1호
• /
• pp.64-73
• /
• 2011

Oil price prediction is an important issue for the regulators of the government and the related industries. When employing the time series techniques for prediction, however, it becomes difficult and challenging since the behavior of the series of oil prices is dominated by quantitatively unexplained irregular external factors, e.g., supply- or demand-side shocks, political conflicts specific to events in the Middle East, and direct or indirect influences from other global economical indices, etc. Identifying and quantifying the relationship between oil price and those external factors may provide more relevant prediction than attempting to unclose the underlying structure of the series itself. Technically, this implies the prediction is to be based on the vectoral data on the degrees of the relationship rather than the series data. This paper proposes a novel method for time series prediction of using Semi-Supervised Learning that was originally designed only for the vector types of data. First, several time series of oil prices and other economical indices are transformed into the multiple dimensional vectors by the various types of technical indicators and the diverse combination of the indicator-specific hyper-parameters. Then, to avoid the curse of dimensionality and redundancy among the dimensions, the wellknown feature extraction techniques, PCA and NLPCA, are employed. With the extracted features, a timepointspecific similarity matrix of oil prices and other economical indices is built and finally, Semi-Supervised Learning generates one-timepoint-ahead prediction. The series of crude oil prices of West Texas Intermediate (WTI) was used to verify the proposed method, and the experiments showed promising results : 0.86 of the average AUC.

• Structural Engineering and Mechanics
• /
• 제44권1호
• /
• pp.51-60
• /
• 2012

Concrete sleepers are essential components of the conventional railway. As support elements, sleepers are always subjective to a variety of time-dependent loads attributable to the train operations, either wheel or rail abnormalities. It has been observed that the sleepers may deteriorate due to these loads, inducing the formation of hairline cracks. There are two areas along the sleepers that are more prone to crack: the central and the rail seat sections. Several non-destructive methods have been developed to identify failures in structures. Health monitoring techniques are based on vibration responses measurements, which help engineers to identify the vibration-based damage or remotely monitor the sleeper health. In the present paper, the dynamic effects of the cracks in the vibration signatures of the railway pre-stressed concrete sleepers are investigated. The experimental modal analysis has been used to evaluate the modal bending changes in the vibration characteristics of the sleepers, differentiating between the central and the rail seat locations of the cracks. Modal parameters changes of the 'healthy' and cracked sleepers have been highlighted in terms of natural frequencies and modal damping. The paper concludes with a discussion of the most suitable failure indicator and it defines the vibration signatures of intact, central cracked and rail seat cracked sleepers.

• 산업경영시스템학회지
• /
• 제39권2호
• /
• pp.64-71
• /
• 2016

Recently, nanotechnology has grown as one of the leading science technology along with other converging technologies such as biology, information, medicine etc., bringing the continuous investment of the government in nano-related field. However, it is difficult to measure and evaluate the performance of the national research and development programs because of the multidimensional character of the expected outcomes. This study aims to measuring efficiency of the national nanotechnology research and development programs using DEA model. The decision making units are nine nano-related ministries including the Ministry of Science, ICT and Future Planning. The input variables are total expenditure, number of the programs and average expenditure per program. The output variables are science, technology and economic indicator, and the combination of these outputs are respectively measured as seven different DEA cases. The Ministry of Science, ICT and Future was the first efficient ministry in total technical efficiency. Ministry of Agriculture, Food and Rural Affairs and the Ministry of Food and Drug Safety were efficient in pure technical efficiency, when the Ministry of Commerce Industry and Energy took the first in the scale efficiency. The program efficiency was affected by organizational characteristics such as the institution's scale, the concentration of the research paper or the patent, technology transfer or the commercialization. The result of this study could be utilized in development of the policy in the nanotechnology and the related field. Furthermore, it could be applied for the modification of expenditure management or the adjustment of the research and development programs' input and output scale for each ministry.

• 한국인터넷방송통신학회논문지
• /
• 제17권5호
• /
• pp.111-117
• /
• 2017

본 논문은 능력중심사회를 구현하고 산업현장과 교육, 훈련, 자격제도의 불일치 문제를 해결하기 위한 국가직무능력표준과 부합되어 취업현장에서 발생하는 재교육과 사회적비용등의 근본적 문제점을 해결하기 위한 인프라로 한국형 NCS직업기초능력평가시스템을 설계하고 부가적으로 시스템의 기술성능 표준지표를 개발하여 한국형 NCS직업기초능력시스템의 해외진출 성과를 창출하기 위한 연구이다. NCS직업기초능력시스템은 컴퓨터, 태블릿PC, 스마트폰등의 멀티디바이스에 적절한 반응형으로 개발되어지며 글로벌 보급을 위하여 연계대상 운영체제, 인터페이스 프로토콜, 패킷포맷, 암호화, Class Component, 동시접속 수, 감독관-수검자 응답속도, 서버-관리자 응답속도, 응시답안 자동복구 속도, 실시간 답안전송 속도의 국제적 표준에 부합되는 10개 성능평가 지표를 설계하고 개발한다.

• 한국IT서비스학회지
• /
• 제19권2호
• /
• pp.109-124
• /
• 2020

Facing the 4^th Industrial Revolution era, researches on artificial intelligence have become active and attempts have been made to apply machine learning in various fields. In the field of finance, Robo Advisor service, which analyze the market, make investment decisions and allocate assets instead of people, are rapidly expanding. The stock price prediction using the machine learning that has been carried out to date is mainly based on the prediction of the market index such as KOSPI, and utilizes technical data that is fundamental index or price derivative index using financial statement. However, most researches have proceeded without any explicit verification of the prediction rate of the learning data. In this study, we conducted an experiment to determine the degree of market prediction ability of basic indicators, technical indicators, and system risk indicators (AR) used in stock price prediction. First, we set the core parameters for each financial indicator and define the objective function reflecting the return and volatility. Then, an experiment was performed to extract the sample from the distribution of each parameter by the Markov chain Monte Carlo (MCMC) method and to find the optimum value to maximize the objective function. Since Robo Advisor is a commodity that trades financial instruments such as stocks and funds, it can not be utilized only by forecasting the market index. The sample for this experiment is data of 17 years of 1,500 stocks that have been listed in Korea for more than 5 years after listing. As a result of the experiment, it was possible to establish a meaningful trading strategy that exceeds the market return. This study can be utilized as a basis for the development of Robo Advisor products in that it includes a large proportion of listed stocks in Korea, rather than an experiment on a single index, and verifies market predictability of various financial indicators.

• 펄프종이기술
• /
• 제47권6호
• /
• pp.106-112
• /
• 2015

Many researchers have proposed analytical methods to measure the adsorption of di-sulpho fluorescent whitening agents (D-FWAs), but practical methods for D-FWA utilization in an actual paper mill have not been established. In particular, the D-FWA adsorption behavior must be monitored in paper mills to ensure the effective use of D-FWAs. This study used the zeta-potential of pulps as an indicator of the adsorption behavior of a D-FWA. We identified the relationship between the actual adsorption of the D-FWA and the zeta-potential of the pulps as a function of D-FWA addition. zeta-potential measurements were then used to analyze the D-FWA adsorption behavior under different conditions of pulp type, conductivity, and pH. The actual adsorption of a D-FWA was proportional to the ${\Delta}zeta-potential$ of the pulps (i.e., the difference between the zeta-potential of a pulp containing no D-FWA and one containing the D-FWA). The ${\Delta}zeta-potential$ of the pulps was therefore adopted for adsorption analysis. A higher adsorption of the D-FWA was observed onto Hw-BKP than onto Sw-BKP because of the shorter fiber length and higher fines content of Hw-BKP. A high conductivity and an acidic pH decreased the D-FWA adsorption because of direct effects of high ion concentrations and low pH on the D-FWA solubility. Therefore, a D-FWA must be added to Hw-BKP under low conductivity conditions and at neutral or alkaline pH to optimize the D-FWA adsorption.