• Journal of Korean Society of Disaster and Security
• /
• v.14 no.3
• /
• pp.17-27
• /
• 2021

In this study, classification models were built using machine learning techniques that can classify the soil creep risk into three classes from A to C (A: risk, B: moderate, C: good). A total of six machine learning techniques were used: K-Nearest Neighbor, Support Vector Machine, Logistic Regression, Decision Tree, Random Forest, and Extreme Gradient Boosting and then their classification accuracy was analyzed using the nationwide soil creep field survey data in 2019 and 2020. As a result of classification accuracy analysis, all six methods showed excellent accuracy of 0.9 or more. The methods where numerical data were applied for data training showed better performance than the methods based on character data of field survey evaluation table. Moreover, the methods learned with the data group (R1~R4) reflecting the expert opinion had higher accuracy than the field survey evaluation score data group (C1~C4). The machine learning can be used as a tool for prediction of soil creep if high-quality data are continuously secured and updated in the future.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.4
• /
• pp.1-9
• /
• 2021

In this paper, we propose a training algorithm of support vector machine (SVM) with a sensitive variable. Although machine learning models enable automatic decision making in the real world applications, regulations prohibit sensitive information from being used to protect privacy. In particular, the privacy protection of the legally protected attributes such as race, gender, and disability is compulsory. We present an efficient least square SVM (LSSVM) training algorithm using a fully homomorphic encryption (FHE) to protect a partial sensitive attribute. Our framework posits that data owner has both non-sensitive attributes and a sensitive attribute while machine learning service provider (MLSP) can get non-sensitive attributes and an encrypted sensitive attribute. As a result, data owner can obtain the encrypted model parameters without exposing their sensitive information to MLSP. In the inference phase, both non-sensitive attributes and a sensitive attribute are encrypted, and all computations should be conducted on encrypted domain. Through the experiments on real data, we identify that our proposed method enables to implement privacy-preserving sensitive LSSVM with FHE that has comparable performance with the original LSSVM algorithm. In addition, we demonstrate that the efficient sensitive LSSVM with FHE significantly improves the computational cost with a small degradation of performance.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.5
• /
• pp.21-32
• /
• 2024

This paper proposes a smart safety system that combines low-cost CW(Continuous Wave) radar and IMU sensors to enhance blind spots that pose safety risks to workers in industrial manufacturing environments. The system employs a 24 GHz radar and a 6-axis IMU sensor to detect worker movements and utilizes a machine learning model to recognize worker situations in vibrating manufacturing sites. The ensemble boosting tree-based model achieved over 92.8% worker detection accuracy, demonstrating its effectiveness in improving safety in industrial settings.

• Journal of Korean Society of Disaster and Security
• /
• v.13 no.2
• /
• pp.27-38
• /
• 2020

Climate variations have become worse and diversified recently, which caused catastrophic disasters for our communities and ecosystem including economic property damages in Korea. Heat wave of summer season is one of causes for such damages of which outbreak tends to increase recently. Related short-term forecasting information has been provided by the Korea Meteorological Administration based on results from numerical forecasting model. As the study area, the ◯◯ province was selected because of the highest mortality rate in Korea for the past 15 years (1998~2012). When comparing the forecasted temperatures with field measurements, it showed RMSE of 1.57℃ and RMSE of 1.96℃ was calculated when only comparing the data corresponding to the observed value of 33℃ or higher. The forecasting process would take at least about 3~4 hours to provide the 4 hours advanced forecasting information. Therefore, this study proposes a methodology for temperature prediction using LSTM considering the short prediction time and the adequate accuracy. As a result of 4 hour temperature prediction using this approach, RMSE of 1.71℃ was occurred. When comparing only the observed value of 33℃ or higher, RMSE of 1.39℃ was obtained. Even the numerical prediction model of the whole range of errors is relatively smaller, but the accuracy of prediction of the machine learning model is higher for above 33℃. In addition, it took an average of 9 minutes and 26 seconds to provide temperature information using this approach. It would be necessary to study for wider spatial range or different province with proper data set in near future.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2001.06a
• /
• pp.383-389
• /
• 2001

된 연구에서 대상으로 하고 있는 LG 정유 광양항 제품부두는 7 선석(Berth)에 재화중량(DWT) 300톤에서 48000 톤의 선박까지 다양한 선박이 이용하고 있으며, 해상의 기상상태에 따른 선박 입출향 통제 지침 설정이 어렵고, 현재 사용하고 있는 지침의 근거가 명확하지 않아 현재의 부두 운영이 비효율적이거나 안전성이 결여되어 있다고 할 수 있다. 따라서 이를 개선하기 위한 합리적인 부두운영 제한조건 개발이 절실히 요구되었다. 본 논문에서는 대상 부두의 특성, 대상 선박의 특성, 하중상태, 선박 운항자의 특성 등을 고려하여 해상/기상 상황(바람, 조류 및 파랑)에 따른 부두 입출항 가능 여부를 정량적으로 판단하고, 안전성 향상 방안을 제시할 수 있는 의사결정 시스템을 개발하고 5번, 7번 선석을 대상으로 이를 검증하였다. 여기서는 입출항 여부를 정량적으로 판단하여 결과를 제시하기 위해서 유전적 프로그래밍(Genetic Programming)을 이용한 기계학습 방법을 이용하였으며, GP의 방대한 계산량을 줄이기 위한 가중 선형 연상 기억(Weighted Linear Associative Memory: WLAM) 방법의 도입 및 전역 최적점을 쉽게 찾기 위한 Group of Additive Genetic Programming Trees(GAGPT)를 도입함으로써 학습 성능을 개선하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.384-384
• /
• 2022

과거의 우리나라 대설피해 양상을 살펴보면 지역적으로 집중되어 피해가 발생하는 것이 특징이다. 그러나 현재는 전국적으로 대설피해가 가중되는 추세이며, 이에 따라 대설피해에 대비 가능한 대책의 강구가 필요한 실정이다. 그러나 피해 발생 시 정확한 피해 예측으로 사전에 재난을 대비가 가능한 수준의 연구는 미흡한 실정이다. 따라서 본 연구에서는 다양한 통계기법과 기계학습 기법을 이용하여 대설로 인해 발생한 피해액을 개략적으로 예측이 가능한 모형을 개발하고자 하였다. 대설피해액 예측 모형은 다중회귀분석, 서포트 벡터 머신, 인공신경망 기법, 랜덤포레스트 기법을 이용하여 총 4가지 기법으로 개발하였으며, 독립변수로 사회·경제적 요소, 기상요소를 사용하였고, 종속변수로는 1994년부터 2020년까지 발생한 대설피해 이력의 대설피해액을 사용하였다. 결과적으로 4가지 예측 모형의 예측력 검증 및 기법 간의 예측력을 비교하여 개발한 모형의 적용성을 검토하였다. 본 연구 결과에서 제시한 모형의 개선방안 및 업데이트 방안을 참고하여 후속 연구가 진행된다면 미래에 전국적으로 확대될 대설피해에 대한 대비가 가능할 것으로 기대되며 복구비 및 예방비 투자의 지역적 우선순위를 분석하여 선제적인 대비가 가능할 것으로 판단된다.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.4
• /
• pp.19-27
• /
• 2020

In this paper, we propose a machine learning method for diagnosing the failure of a gas pressure regulator. Originally, when implementing a machine learning model for detecting abnormal operation of a facility, it is common to install sensors to collect data. However, failure of a gas pressure regulator can lead to fatal safety problems, so that installing an additional sensor on a gas pressure regulator is not simple. In this paper, we propose various machine learning approach for diagnosing the abnormal operation of a gas pressure regulator with only the flow rate and gas pressure data collected from a gas pressure regulator itself. Since the fault data of a gas pressure regulator is not enough, the model is trained in all classes by applying the over-sampling method. The classification model was implemented using Gradient boosting, 1D Convolutional Neural Networks, and LSTM algorithm, and gradient boosting model showed the best performance among classification models with 99.975% accuracy.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.3
• /
• pp.7-14
• /
• 2021

The emergence of the automated smart grid has become an essential device for responding to these problems and is bringing progress toward a smart grid-based society. Smart grid is a new paradigm that enables two-way communication between electricity suppliers and consumers. Smart grids have emerged due to engineers' initiatives to make the power grid more stable, reliable, efficient and safe. Smart grids create opportunities for electricity consumers to play a greater role in electricity use and motivate them to use electricity wisely and efficiently. Therefore, this study focuses on power demand management through machine learning. In relation to demand forecasting using machine learning, various machine learning models are currently introduced and applied, and a systematic approach is required. In particular, the GP learning model has advantages over other learning models in terms of general consumption prediction and data visualization, but is strongly influenced by data independence when it comes to prediction of smart meter data.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.10
• /
• pp.38-47
• /
• 1998

The syntactic analysis for the practical machine translation should be able to analyze a long sentence, but the long sentence analysis is a critical problem because of its high analysis complexity. In this paper a sentence segmentation method is proposed for an efficient analysis of a long sentence and the method of determining optimal sentence segmentation positions using statistical information and genetic learning is introduced. It consists of two modules: (1) decomposable position determination which uses lexical contextual constraints acquired from a training data tagged with segmentation positions. (2) segmentation position selection by the selection function of which the weights of parameters are determined through genetic learning, which selects safe segmentation positions with enhancing the analysis efficiency as much as possible. The safe segmentation by the proposed sentence segmentation method and the efficiency enhancement of the analysis are presented through experiments.

• Journal of Convergence for Information Technology
• /
• v.11 no.6
• /
• pp.270-281
• /
• 2021

With the development of information technology, the size of insurance fraud is increasing rapidly every year, and the method is being organized and advanced in conspiracy. Although various forms of prediction models are being studied to predict and detect this, insurance-related information is highly sensitive, which poses a high risk of sharing and access and has many legal or technical constraints. In this paper, we propose a machine learning insurance fraud prediction model based on blockchain, one of the most popular technologies with the recent advent of the Fourth Industrial Revolution. We utilize blockchain technology to realize a safe and trusted insurance information sharing system, apply the theory of social relationship analysis for more efficient and accurate fraud prediction, and propose machine learning fraud prediction patterns in four stages. Claims with high probability of fraud have the effect of being detected at a higher prediction rate at an earlier stage, and claims with low probability are applied differentially for post-reference management. The core mechanism of the proposed model has been verified by constructing an Ethereum local network, requiring more sophisticated performance evaluations in the future.