• 응용통계연구
• /
• 제32권6호
• /
• pp.795-835
• /
• 2019

최근 다양한 분야에서 '빅데이터'가 생성되었다. 많은 기업들은 인공지능(AI)을 기반으로 빅데이터 분석이 가능한 시스템을 구축하여 이익 창출을 시도하고 있다. 인공지능 기술을 접목함으로써 방대한 양의 데이터를 효율적으로 분석하고 효과적으로 활용하는 것은 점점 더 중요해지고 있다. 특히 재무, 조달, 생산 및 마케팅과 같은 다양한 분야에서 국가 및 기업 경영 관리에있어 최소의 오차와 최대의 정확도를 갖춘 수요예측은 절대적으로 중요한 요소이다. 이때 각 분야의 수요패턴을 고려한 적절한 모델을 적용하는 것이 중요하다. 전통적으로 쓰이는 시계열모델이나 회귀모델로도 비대해진 실제 데이터의 복잡한 비선형적인 패턴을 분석할 수 있다. 그러나 다양한 비선형 모델들 중에서 적절한 모델을 선택하는 것은 사전 지식 없이는 어려운 일이다. 최근에는 인공지능 기반의 기법들인 머신러닝이나 딥러닝 기법을 중심으로 이루어진 연구들이 이를 극복할 수 있음을 증명하고 있다. 뿐만 아니라 정형데이터와 이미지나 텍스트의 비정형 데이터 분석을 통한 수요예측도 높은 정확도를 갖춘 결과를 보이고 있다. 따라서 본 연구에서는 수요예측이 비교적 활발하게 일어나는 중요한 분야들을 나누어 설명하였다. 그리고 각 분야별로 갖는 특징적인 성격을 고려한 인공지능 기반의 수요예측 기법에 대해 머신러닝과 딥러닝 기법으로 나누어 소개하였다.

• 한국측량학회지
• /
• 제35권5호
• /
• pp.349-356
• /
• 2017

In this study, we propose a method to investigate topographic characteristics by applying machine learning which is an artificial intelligence analysis method based on the spatial data constructed using UAV and the training data created through spatial analysis. This method provides an alternative to the subjective judgment and accuracy of spatial data, which is a problem of existing topographic characteristics survey for officially assessed land price. The analysis method of this study is expected to improve the problems of topographic characteristics survey method of existing field researchers and contribute to more accurate decision of officially assessed land price by providing more objective land survey method.

• Structural Engineering and Mechanics
• /
• 제88권6호
• /
• pp.535-549
• /
• 2023

Deep extreme learning machine (DELM) and multi-verse optimization algorithms (MVO) are hybridized for designing an optimal and adaptive control framework for uncertain buildings. In this approach, first, a robust model predictive control (RMPC) scheme is developed to handle the problem uncertainty. The optimality and adaptivity of the proposed controller are provided by the optimal determination of the tunning weights of the linear programming (LP) cost function for clustered external loads using the MVO. The final control policy is achieved by collecting the clustered data and training them by DELM. The efficiency of the introduced control scheme is demonstrated by the numerical simulation of a ten-story benchmark building subjected to earthquake excitations. The results represent the capability of the proposed framework compared to robust MPC (RMPC), conventional MPC (CMPC), and conventional DELM algorithms in structural motion control.

• 시스템엔지니어링학술지
• /
• 제19권2호
• /
• pp.18-31
• /
• 2023

Exploring artificial intelligence and machine learning for nuclear safety has witnessed increased interest in recent years. To contribute to this area of research, a machine learning model capable of accurately predicting nuclear power plant response with minimal computational cost is proposed. To develop a robust machine learning model, the Best Estimate Plus Uncertainty (BEPU) approach was used to generate a database to train three models and select the best of the three. The BEPU analysis was performed by coupling Dakota platform with the best estimate thermal hydraulics code RELAP/SCDAPSIM/MOD 3.4. The Code Scaling Applicability and Uncertainty approach was adopted, along with Wilks' theorem to obtain a statistically representative sample that satisfies the USNRC 95/95 rule with 95% probability and 95% confidence level. The generated database was used to train three models based on Recurrent Neural Networks; specifically, Long Short-Term Memory, Gated Recurrent Unit, and a hybrid model with Long Short-Term Memory coupled to Convolutional Neural Network. In this paper, the System Engineering approach was utilized to identify requirements, stakeholders, and functional and physical architecture to develop this project and ensure success in verification and validation activities necessary to ensure the efficient development of ML meta-models capable of predicting of the nuclear power plant response.

• 한국산업융합학회 논문집
• /
• 제27권1호
• /
• pp.87-95
• /
• 2024

This study is on modulo scheduling algorithms for multicore processor in machine learning applications. Machine learning algorithms are designed to perform a large amount of operations such as vectors and matrices in order to quickly process large amounts of data stream. To support such large amounts of computations, processor architectures to support applications such as artificial intelligence, neural networks, and machine learning are designed in the form of parallel processing such as multicore. To effectively utilize these multi-core hardware resources, various compiler techniques are being used and studied. In this study, among these compiler techniques, we analyzed the modular scheduler, which is especially important in one core's computation pipeline. This paper looked at and compared the iterative modular scheduler and the swing modular scheduler, which are the most widely used and studied. As a result, both schedulers provided similar performance results, and when measuring register pressure as an indicator, it was confirmed that the swing modulo scheduler provided slightly better performance. In this study, a technique that divides recurrence edge is proposed to improve the minimum initiation interval of the modulo schedulers.

• Steel and Composite Structures
• /
• 제49권4호
• /
• pp.419-430
• /
• 2023

This paper is aimed at investigating the effect of multiple longitudinal stiffeners on the patch loading resistance of slender steel plate girders. Firstly, a numerical study is conducted through geometrically and materially nonlinear analysis with imperfections included (GMNIA), the model is validated with experimental results taken from the literature. The structural responses of girders with multiple longitudinal stiffeners are compared to the one of girders with a single longitudinal stiffener. Thereafter, a patch loading resistance model is developed through machine learning (ML) using symbolic regression (SR). An extensive numerical dataset covering a wide range of bridge girder geometries is employed to fit the resistance model using SR. Finally, the performance of the SR prediction model is evaluated by comparison of the resistances predicted using available formulae from the literature.

• Computers and Concrete
• /
• 제28권6호
• /
• pp.533-547
• /
• 2021

Calculating the shear capacity of slender reinforced concrete beams without shear reinforcement was the subject of numerous studies, where the eternal problem of developing a single relationship that will be able to predict the expected shear capacity is still present. Using experimental results to extrapolate formulae was so far the main approach for solving this problem, whereas in the last two decades different research studies attempted to use artificial intelligence algorithms and available data sets of experimentally tested beams to develop new models that would demonstrate improved prediction capabilities. Given the limited number of available experimental databases, these studies were numerically restrained, unable to holistically address this problem. In this manuscript, a new approach is proposed where a numerically generated database is used to train machine-learning algorithms and develop an improved model for predicting the shear capacity of slender concrete beams reinforced only with longitudinal rebars. Finally, the proposed predictive model was validated through the use of an available ACI database that was developed by using experimental results on physical reinforced concrete beam specimens without shear and compressive reinforcement. For the first time, a numerically generated database was used to train a model for computing the shear capacity of slender concrete beams without stirrups and was found to have improved predictive abilities compared to the corresponding ACI equations. According to the analysis performed in this research work, it is deemed necessary to further enrich the current numerically generated database with additional data to further improve the dataset used for training and extrapolation. Finally, future research work foresees the study of beams with stirrups and deep beams for the development of improved predictive models.

• 한국수학교육학회지시리즈D:수학교육연구
• /
• 제26권4호
• /
• pp.333-349
• /
• 2023

Recently, AI has become a crucial tool in mathematics education due to advances in machine learning and deep learning. Considering the importance of AI, examining teachers' beliefs about AI in mathematics education (AIME) is crucial, as these beliefs affect their instruction and student learning experiences. The present study developed a scale to measure preservice teachers' (PST) beliefs about AIME through factor analysis and rigorous reliability and validity analyses. The study analyzed 202 PST's data and developed a scale comprising three factors and 11 items. The first factor gauges PSTs' beliefs regarding their roles in using AI for mathematics education (4 items), the second factor assesses PSTs' beliefs about using AI for mathematics teaching (3 items), and the third factor explores PSTs' beliefs about AI for mathematics learning (4 items). Moreover, the outcomes of confirmatory factor analysis affirm that the three-factor model outperforms other models (a one-factor or a two-factor model). These findings are in line with previous scales examining mathematics teacher beliefs, reinforcing the notion that such beliefs are multifaceted and developed through diverse experiences. Descriptive analysis reveals that overall PSTs exhibit positive beliefs about AIME. However, they show relatively lower levels of beliefs about their roles in using AI for mathematics education. Practical and theoretical implications are discussed.

• 반도체디스플레이기술학회지
• /
• 제21권3호
• /
• pp.110-113
• /
• 2022

Non-playable game character plays an important role in improving the concentration of the game and the interest of the user, and recently implementation of NPC with reinforcement learning has been in the spotlight. In this paper, we estimate an AI target tracking method via reinforcement learning, and implement an AI-based tracking agency of specific target object with avoiding traps through Unity ML-Agents Toolkit. The implementation is built in Unity game engine, and simulations are conducted through a number of experiments. The experimental results show that outstanding performance of the tracking target with avoiding traps is shown with good enough results.

• Nuclear Engineering and Technology
• /
• 제53권2호
• /
• pp.522-531
• /
• 2021

We describe a new deep learning model - Convolutional eXtreme Gradient Boosting (ConvXGB) for classification problems based on convolutional neural nets and Chen et al.'s XGBoost. As well as image data, ConvXGB also supports the general classification problems, with a data preprocessing module. ConvXGB consists of several stacked convolutional layers to learn the features of the input and is able to learn features automatically, followed by XGBoost in the last layer for predicting the class labels. The ConvXGB model is simplified by reducing the number of parameters under appropriate conditions, since it is not necessary re-adjust the weight values in a back propagation cycle. Experiments on several data sets from UCL Repository, including images and general data sets, showed that our model handled the classification problems, for all the tested data sets, slightly better than CNN and XGBoost alone and was sometimes significantly better.