• Nuclear Engineering and Technology
• /
• v.55 no.2
• /
• pp.603-622
• /
• 2023

Human error (HE) is an important concern in safety-critical systems such as nuclear power plants (NPPs). HE has played a role in many accidents and outage incidents in NPPs. Despite the increased automation in NPPs, HE remains unavoidable. Hence, the need for HE detection is as important as HE prevention efforts. In NPPs, HE is rather rare. Hence, anomaly detection, a widely used machine learning technique for detecting rare anomalous instances, can be repurposed to detect potential HE. In this study, we develop an unsupervised anomaly detection technique based on generative adversarial networks (GANs) to detect anomalies in manually collected surveillance data in NPPs. More specifically, our GAN is trained to detect mismatches between automatically recorded sensor data and manually collected surveillance data, and hence, identify anomalous instances that can be attributed to HE. We test our GAN on both a real-world dataset and an external dataset obtained from a testbed, and we benchmark our results against state-of-the-art unsupervised anomaly detection algorithms, including one-class support vector machine and isolation forest. Our results show that the proposed GAN provides improved anomaly detection performance. Our study is promising for the future development of artificial intelligence based HE detection systems.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.804-806
• /
• 2020

딥러닝(deep learning)은 기계학습 알고리즘 중 가장 널리 활용되고 있는 알고리즘이다. 딥러닝 기술은 산업, 과학, 국방 및 공공 부문을 비롯하여 거의 모든 분야에서 폭넓게 확산되고 있다. 그러나 기계학습 기술에 대한 이해와 프로그래밍 지식이 부족할 경우 자유롭게 활용하는 데는 제약이 따르고 있으며 빅데이터를 활용하여 일반 이용자들이 직접 분산 학습 모형을 개발하고 배포하는 데 어려움이 발생하고 있다. 이러한 요구를 충족시키기 위해 딥러닝 프레임워크의 저수준 API를 추상화하여 고수준 분석과 분산 딥러닝을 지원하고 일반 이용자들이 실무적으로 복잡한 딥러닝 기술을 활용할 수 있는 기술을 개발하였다. 플랫폼 개발과 함께 중요하게 고려해야 하는 요소 중 하나로 플랫폼의 배포와 확장성 역시 고려되어야 한다. 본 플랫폼은 조직 내 계산 자원을 이용하여 플랫폼을 배포할 수 있으며 상용 클라우드 서비스와 연동하여 배포할 수 있도록 설계됨에 따라 환경의 제약 없이 유연한 서비스 제공이 가능하다.

• Journal of Information Processing Systems
• /
• v.9 no.4
• /
• pp.633-650
• /
• 2013

All the imputation techniques proposed so far in literature for data imputation are offline techniques as they require a number of iterations to learn the characteristics of data during training and they also consume a lot of computational time. Hence, these techniques are not suitable for applications that require the imputation to be performed on demand and near real-time. The paper proposes a computational intelligence based architecture for online data imputation and extended versions of an existing offline data imputation method as well. The proposed online imputation technique has 2 stages. In stage 1, Evolving Clustering Method (ECM) is used to replace the missing values with cluster centers, as part of the local learning strategy. Stage 2 refines the resultant approximate values using a General Regression Neural Network (GRNN) as part of the global approximation strategy. We also propose extended versions of an existing offline imputation technique. The offline imputation techniques employ K-Means or K-Medoids and Multi Layer Perceptron (MLP)or GRNN in Stage-1and Stage-2respectively. Several experiments were conducted on 8benchmark datasets and 4 bank related datasets to assess the effectiveness of the proposed online and offline imputation techniques. In terms of Mean Absolute Percentage Error (MAPE), the results indicate that the difference between the proposed best offline imputation method viz., K-Medoids+GRNN and the proposed online imputation method viz., ECM+GRNN is statistically insignificant at a 1% level of significance. Consequently, the proposed online technique, being less expensive and faster, can be employed for imputation instead of the existing and proposed offline imputation techniques. This is the significant outcome of the study. Furthermore, GRNN in stage-2 uniformly reduced MAPE values in both offline and online imputation methods on all datasets.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.6
• /
• pp.1957-1980
• /
• 2021

The significant aim of this research has always been to evaluate the mechanism for efficient and inherently aware usage of vitality in-home devices, thus improving the information of smart metering systems with regard to the usage of selected homes and the time of use. Advances in information processing are commonly used to quantify gigantic building activity data steps to boost the activity efficiency of the building energy systems. Here, some smart data mining models are offered to measure, and predict the time series for energy in order to expose different ephemeral principles for using energy. Such considerations illustrate the use of machines in relation to time, such as day hour, time of day, week, month and year relationships within a family unit, which are key components in gathering and separating the effect of consumers behaviors in the use of energy and their pattern of energy prediction. It is necessary to determine the multiple relations through the usage of different appliances from simultaneous information flows. In comparison, specific relations among interval-based instances where multiple appliances use continue for certain duration are difficult to determine. In order to resolve these difficulties, an unsupervised energy time-series data clustering and a frequent pattern mining study as well as a deep learning technique for estimating energy use were presented. A broad test using true data sets that are rich in smart meter data were conducted. The exact results of the appliance designs that were recognized by the proposed model were filled out by Deep Convolutional Neural Networks (CNN) and Recurrent Neural Networks (LSTM and GRU) at each stage, with consolidated accuracy of 94.79%, 97.99%, 99.61%, for 25%, 50%, and 75%, respectively.

• The Journal of Society for e-Business Studies
• /
• v.25 no.2
• /
• pp.49-63
• /
• 2020

The accurate prediction of box office in the early stage is crucial for film industry to make better managerial decision. With aims to improve the prediction performance, the purpose of this paper is to evaluate the use of machine learning methods. We tested both classification and regression based methods including k-NN, SVM and Random Forest. We first evaluate input variables, which show that reputation-related information generated during the first two-week period after release is significant. Prediction test results show that regression based methods provides lower prediction error, and Random Forest particularly outperforms other machine learning methods. Regression based method has better prediction power when films have small box office earnings. On the other hand, classification based method works better for predicting large box office earnings.

• Journal of the Korean GEO-environmental Society
• /
• v.22 no.10
• /
• pp.5-12
• /
• 2021

The deterioration of facilities is an unavoidable phenomenon. For the management of aging facilities, cracks can be detected and tracked, and the condition of the facilities can be indirectly inferred. Therefore, crack detection plays a crucial role in the management of aged facilities. Conventional maintenances are conducted using the crack detection results. For example, maintenance activities to prevent further deterioration can be performed. However, currently, most crack detection relies only on human judgment, so if the area of the facility is large, cost and time are excessively used, and different judgment results may occur depending on the expert's competence, it causes reliability problems. This paper proposes a concrete crack detection framework based on machine learning to overcome these limitations. Fully automated concrete crack detection was possible through the proposed framework, which showed a high accuracy of 96%. It is expected that effective and efficient management will be possible through the proposed framework in this paper.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.5
• /
• pp.109-126
• /
• 2023

Early detection of forest fires is essential in preventing large-scale forest fires. Predicting forest fires serves as a vital early detection method, leading to various related studies. However, many previous studies focused solely on climate and geographic factors, overlooking human factors, which significantly contribute to forest fires. This study aims to develop forest fire prediction models that take into account human, weather and geographical factors. This study conducted a comparative analysis of four machine learning models alongside the logistic regression model, using forest fire data from Gangwon-do spanning 2003 to 2020. The results indicate that XG Boost models performed the best (AUC=0.925), closely followed by Random Forest (AUC=0.920), both of which are machine learning techniques. Lastly, the study analyzed the relative importance of various factors through permutation feature importance analysis to derive operational insights. While meteorological factors showed a greater impact compared to human factors, various human factors were also found to be significant.

• Information Systems Review
• /
• v.20 no.4
• /
• pp.25-41
• /
• 2018

Applying predictive analytics to enterprise processes is an effective way to reduce operation costs and enhance productivity. Accordingly, the ability to predict business processes and performance indicators are regarded as a core capability. Recently, several works have predicted processes using deep learning in the form of recurrent neural networks (RNN). In particular, the approach of predicting the next step of activity using static or dynamic RNN has excellent results. However, few studies have given attention to applying deep learning in the form of dynamic RNN to predictions of process performance indicators. To fill this knowledge gap, the study developed an approach to using process mining and dynamic RNN. By utilizing actual data from a large domestic company, it has applied the suggested approach in estimating timely stocking in purchasing process, which is an important indicator of the process. The analytic methods and results of this study were presented and some implications and limitations are also discussed.

• Journal of Navigation and Port Research
• /
• v.42 no.5
• /
• pp.341-346
• /
• 2018

Supramax bulk carriers cover a wide range of ocean transportation requirements, from major to minor bulk cargoes. Market forecasting for this segment has posed a challenge to researchers, due to complexity involved, on the demand side of the forecasting model. This paper addresses this issue by using technical indicators as input features, instead of complicated supply-demand variables. Artificial neural networks (ANN), one of the most popular machine-learning tools, were used to replace classical time-series models. Results revealed that ANN outperformed the benchmark binomial logistic regression model, and predicted direction of the spot market with more than 70% accuracy. Results obtained in this paper, can enable chartering desks to make better short-term chartering decisions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.8
• /
• pp.4090-4102
• /
• 2018

After emerging online communications, text mining and sentiment analysis has been frequently applied into analyzing electronic word-of-mouth. This study aims to develop a domain-specific lexicon of sentiment analysis to predict box office success in Korea film market and validate the feasibility of the lexicon. Natural language processing, a machine learning algorithm, and a lexicon-based sentiment classification method are employed. To create a movie domain sentiment lexicon, 233,631 reviews of 147 movies with popularity ratings is collected by a XML crawling package in R program. We accomplished 81.69% accuracy in sentiment classification by the Korean sentiment dictionary including 706 negative words and 617 positive words. The result showed a stronger positive relationship with box office success and consumers' sentiment as well as a significant positive effect in the linear regression for the predicting model. In addition, it reveals emotion in the user-generated content can be a more accurate clue to predict business success.