• 재무관리연구
• /
• 제19권2호
• /
• pp.211-231
• /
• 2002

본 연구의 목적은 VECM(Vector Error Correction Model)과 인공지능모형(Artificial Neural Networks)을 이용하여 우리나라 증권시장과 거시경제 변수들과의 장기적 관계에 대한 설명력을 비교해보고자 함에 있다. VECM이 APT(Arbitrage Pricing Theory)에 기초를 둔 선형동학모형이라고 한다면, 인공지능모형은 비모수적 비선형모형이라는 점에서, 두 방법론의 분석결과를 직접 비판하는 것은 의미있는 연구라고 할 수 있다. 인공지능모형을 주로 활용하는 선행연구들에 의하면, 증권시장은 시장의 특이패턴들로 인해 계량경제학적 접근인 선형 모형보다는 인공지능모형을 통해 증권시장의 움직임을 설명하고 예측하는 것이 더 바람직할 수도 있다는 것이다. 따라서, 본 연구에서는 VECM분석에서 자료의 안정성을 검증하고, 공적분 백터를 발견한 이후, 장기적 균형관계의 실증적 분석을 하였다. 그리고, 인공지능모형에서는 delta rule과 Sigmoid 함수를 이용한 GRNN(General Regression Neural Net)과 Back-Propagation등의 방법들을 활용하였다. 이러한 분석결과, Back-Propagation 모형이 다른 모든 모형들보다도 더 우수한 설명력을 보여주고 있었다. 이러한 결과들은 인공지능모형이 동태적인 선형 모형보다도 더 우수한 설명력을 제공할 수 있는 가능성을 보여주고 있었다.

• Neurospine
• /
• 제15권4호
• /
• pp.329-337
• /
• 2018

Objective: Machine learning algorithms excel at leveraging big data to identify complex patterns that can be used to aid in clinical decision-making. The objective of this study is to demonstrate the performance of machine learning models in predicting postoperative complications following anterior cervical discectomy and fusion (ACDF). Methods: Artificial neural network (ANN), logistic regression (LR), support vector machine (SVM), and random forest decision tree (RF) models were trained on a multicenter data set of patients undergoing ACDF to predict surgical complications based on readily available patient data. Following training, these models were compared to the predictive capability of American Society of Anesthesiologists (ASA) physical status classification. Results: A total of 20,879 patients were identified as having undergone ACDF. Following exclusion criteria, patients were divided into 14,615 patients for training and 6,264 for testing data sets. ANN and LR consistently outperformed ASA physical status classification in predicting every complication (p < 0.05). The ANN outperformed LR in predicting venous thromboembolism, wound complication, and mortality (p < 0.05). The SVM and RF models were no better than random chance at predicting any of the postoperative complications (p < 0.05). Conclusion: ANN and LR algorithms outperform ASA physical status classification for predicting individual postoperative complications. Additionally, neural networks have greater sensitivity than LR when predicting mortality and wound complications. With the growing size of medical data, the training of machine learning on these large datasets promises to improve risk prognostication, with the ability of continuously learning making them excellent tools in complex clinical scenarios.

• 대한설비공학회:학술대회논문집
• /
• 대한설비공학회 2009년도 하계학술발표대회 논문집
• /
• pp.940-945
• /
• 2009

Energy planning for hybrid energy system is important to increase the flexibility in the urban community and national energy systems. Expected maximum loads, load profiles and yearly energy demands are important input parameters to plan for the technical and environmental optimal energy system for a planning area. The method for energy demand prediction has been based on artificial neural networks(ANN). The advantage of ANN with respect to the other method is their ability of modeling a multivariable problem given by the complex relationships between the variables. This method can produce 10% of errors hourly load profile from individual building to urban community. As the results of this paper, energy demand prediction system has been developed based on simulink.

• 한국게임학회 논문지
• /
• 제16권2호
• /
• pp.27-40
• /
• 2016

스텔스 게임에서 플레이어의 행동을 예측하는 것은 게임 디자인에 있어서 핵심적인 역할을 한다. 하지만, 플레이어와 게임 환경 간의 상호작용이 실시간으로 일어난다는 점에서 이러한 예측 프로세스를 자동화하는 것은 어려운 문제이다. 본 논문은 동적 환경에서의 스텔스 움직임을 예측하기 위한 강화학습 방법을 소개하며, 이를 위해 Q-learning과 인공신경망이 통합된 형태의 모델이 액션 시뮬레이션을 위한 분류기로 활용된다. 실험 결과들은 이러한 시뮬레이션 에이전트가 동적으로 변하는 주변 상황에 민감하게 반응함을 보여주며, 따라서 게임 레벨 디자이너가 다양한 게임 요소들을 결정하는데 유용함을 보여준다.

• Journal of Electrical Engineering and Technology
• /
• 제10권6호
• /
• pp.2277-2287
• /
• 2015

In recent times, multilevel inverters are given high priority in many large industrial drive applications. However, the reliability of multilevel inverters are mainly affected by the failure of power electronic switches. In this paper, open-switch and short-switch failure of multilevel inverters and its identification using a high performance diagnostic system is discussed. Experimental and simulation studies were carried out on five level cascaded H-Bridge multilevel inverter and its output voltage waveforms were analyzed at different switch fault cases and at different modulation index values. Salient frequency domain features of the output voltage signal were extracted using the discrete wavelet transform multi resolution signal decomposition technique. Real time application of the proposed fault diagnostic system was implemented through the LabVIEW software. Artificial neural network was trained offline using the Matlab software and the resultant network parameters were transferred to LabVIEW real time system. In the proposed system, it is possible to precisely identify the individual faulty switch (may be due to open-switch (or) short-switch failure) of multilevel inverters.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권7호
• /
• pp.2736-2754
• /
• 2015

Android dominates the mobile operating system market, which stimulates the rapid spread of mobile malware. It is quite challenging to detect mobile malware. System call sequence analysis is widely used to identify malware. However, the malware detection accuracy of existing approaches is not satisfactory since they do not consider correlation of system calls in the sequence. In this paper, we propose a new scheme called Artificial Neural Networks (ANNs) on Co-occurrence Matrices Droid (ANNCMDroid), using co-occurrence matrices to mine correlation of system calls. Our key observation is that correlation of system calls is significantly different between malware and benign software, which can be accurately expressed by co-occurrence matrices, and ANNs can effectively identify anomaly in the co-occurrence matrices. Thus at first we calculate co-occurrence matrices from the system call sequences and then convert them into vectors. Finally, these vectors are fed into ANN to detect malware. We demonstrate the effectiveness of ANNCMDroid by real experiments. Experimental results show that only 4 applications among 594 evaluated benign applications are falsely detected as malware, and only 18 applications among 614 evaluated malicious applications are not detected. As a result, ANNCMDroid achieved an F-Score of 0.981878, which is much higher than other methods.

• International Journal of Aeronautical and Space Sciences
• /
• 제18권4호
• /
• pp.651-661
• /
• 2017

A truss-braced wing (TBW) aircraft has recently received increasing attention due to higher aerodynamic efficiency compared to conventional cantilever wing aircraft. For conceptual TBW aircraft design, we developed a propulsion-and-airframe integrated design environment by replacing a semi-empirical turbofan engine model with a thermodynamic cycle-based one built upon the numerical propulsion system simulation (NPSS). The constructed NPSS model benefitted TBW aircraft design study, as it could handle engine installation effects influencing engine fuel efficiency. The NPSS model also contributed to broadening TBW aircraft design space, for it provided turbofan engine design variables involving a technology factor reflecting progress in propulsion technology. To effectively consolidate the NPSS propulsion model with the TBW airframe model, we devised a rapid, approximate substitute of the NPSS model by reduced-order modeling (ROM) to resolve difficulties in model integration. In addition, we formed an artificial neural network (ANN) that associates engine component attributes evaluated by object-oriented weight analysis of turbine engine (WATE++) with engine design variables to determine engine weight and size, both of which bring together the propulsion and airframe system models. Through propulsion-andairframe design space exploration, we optimized TBW aircraft design for fuel saving and revealed that a simple engine model neglecting engine installation effects may overestimate TBW aircraft performance.

• 한국산학기술학회논문지
• /
• 제12권2호
• /
• pp.712-717
• /
• 2011

동적 델타 헤징(Dynamic Delta Hedging)이란 옵션 발행자가 옵션의 만기정산금액(payoff)을 지급하기 위해 주기적으로 델타에 근거한 헤지 포지션을 조절함으로써 옵션의 payoff를 복제하고 옵션 가치변화에 따른 위험을 회피하는 방법이다. 본 연구에서는 헤지에 있어서 주요 변수인 블랙-숄즈의 모형에 의해 산출된 델타의 대체 값을 찾기 위해 기계학습의 일종인 인공신경망 학습을 적용하여 옵션의 만기 시 헤지 비용의 최소화 및 차익 실현을 위한 방법론을 제시하고자 한다. 기초자산의 현재가격, 변동성, 무위험이자율, 만기 등의 시장 상황 변화에 따른 다양한 시나리오에 대한 실험을 통해 본 연구에서 제시하는 방법론의 성능을 분석하고 그 우수성을 보인다.

• Steel and Composite Structures
• /
• 제47권5호
• /
• pp.601-613
• /
• 2023

Numerous methods have been proposed in predicting formability of sheet metals based on microstructural and macro-scale properties of sheets. However, there are limited number of papers on the optimization problem to increase formability of sheet metals. In the present study, we aim to use novel optimization algorithms in neural networks to maximize the formability of sheet metals based on tensile curve and texture of aluminum sheet metals. In this regard, experimental and numerical evaluations of effects of texture and tensile properties are conducted. The texture effects evaluation is performed using Taylor homogenization method. The data obtained from these evaluations are gathered and utilized to train and validate an artificial neural network (ANN) with different optimization methods. Several optimization method including grey wolf algorithm (GWA), chimp optimization algorithm (ChOA) and whale optimization algorithm (WOA) are engaged in the optimization problems. The results demonstrated that in aluminum alloys the most preferable texture is cube texture for the most formable sheets. On the other hand, slight differences in the tensile behavior of the aluminum sheets in other similar conditions impose no significant decreases in the forming limit diagram under stretch loading conditions.

• 반도체디스플레이기술학회지
• /
• 제22권4호
• /
• pp.179-184
• /
• 2023

We propose the modeling methodology of CMOS inverter made of LTPO TFT using a machine learning. LTPO can achieve advantages of LTPS TFT with high electron mobility as a driving TFT and IGZO TFT with low off-current as a switching TFT. However, since the unified model of both LTPS and IGZO TFTs is still lacking, it is necessary to develop a SPICE-compatible compact model to simulate the LTPO current-voltage characteristics. In this work, a generic framework for combining the existing formula of I-V characteristics with artificial neural network is presented. The weight and bias values of ANN for LTPS and IGZO TFTs is obtained and implemented into PSPICE circuit simulator to predict CMOS inverter. This methodology enables efficient modeling for predicting LTPO TFT circuit characteristics.