• 한국안전학회지
• /
• 제38권1호
• /
• pp.1-8
• /
• 2023

It is extremely important to design safety-guaranteed industrial processes because such process determine the ultimate outcomes of industrial activities, including worker safety. Application of artificial intelligence (AI) in industrial safety involves modeling industrial safety systems by using vast amounts of safety-related data, accident prediction, and accident prevention based on predictions. As a preliminary step toward realizing AI-based industrial safety in Korea, this study discusses neural network-based modeling of industrial safety systems. The input variables that are the most discriminatory relative to the output variables of industrial safety processes are selected using two information-theoretic measures, namely entropy and cross entropy. Normalized frequency and severity of industrial accidents are selected as the output variables. Our simulation results confirm the effectiveness of the proposed neural network model and, therefore, the feasibility of extending the model to include more input and output variables.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2008년도 춘계학술대회 논문집
• /
• pp.158-163
• /
• 2008

In inverse synthetic aperture radar (ISAR) imaging, An ISAR autofocusing algorithm is essential to obtain well-focused ISAR images. Traditional methods have relied on the approximation that the phase error due to target motion is a function of the cross-range dimension only. However, in the stepped-frequency radar system, it tends to become a two-dimensional function of both down-range and cross-range, especially when target's movement is very fast and the pulse repetition frequency (PRF) is low. In order to remove the phase error along down-range, this paper proposes a method called SAEM (subarray averaging and entropy minimization) [1] that uses a subarray averaging concept in conjunction with the entropy cost function in order to find target motion parameters, and a novel 2-D optimization technique with the inherent properties of the proposed entropy-based cost function. A well-focused ISAR image can be obtained from the combination of the proposed method and a traditional autofocus algorithm that removes the phase error along the cross-range dimension. The effectiveness of this method is illustrated and analyzed with simulated targets comprised of point scatters.

• 대한전자공학회논문지SP
• /
• 제41권6호
• /
• pp.89-97
• /
• 2004

빛을 반사하거나 방출하는 물체의 형상은 여러 원인에 의해 왜곡된 영상으로 센서에 포착된다. 이러한 왜곡을 제거해 원래 물체의 형상을 추정하는 것을 영상 복원이라고 한다. 영상 복원은 결정론적 방법과 확률론적 방법이 있다. 본 논문에서는 확률론적 방법의 한 종류로서 피셔 인포메이션(Fisher Information)으로부터 유도된 MFI(Minimum Fisher Information)을 이용한 영상 복원을 제안한다. 이는 B. Roy Frieden에 의해 최근에 제안된 신호 추정 방법의 하나이다. MFI을 이용한 복원에서 노이즈 제어 파라미터에 따라 영상 복원의 결과가 어떻게 변화하는지를 조사하였으며, 복원의 정확도에 대한 기준으로 크로스 엔트로피(Kullback-Leibler entropy)를 사용하였다.

• 한국콘텐츠학회논문지
• /
• 제17권1호
• /
• pp.201-212
• /
• 2017

본 연구에서는 변동계수를 이용하여 DEA 모형의 변별력 평가에 적용할 수 있는 새로운 평가기준을 제시하였다. 변별력 평가를 위해 기존 연구에서 제시한 중요도와 본 연구에서 제안한 변동계수를 이용하여 변별력을 분석하였다. 다양한 DEA 모형들 중 변별력 평가를 위해 CCR-DEA, BCC-DEA, entropy, bootstrap, super efficiency, cross efficiency DEA 모형을 선정하고 실증분석을 실시하였다. 모형들의 순위상관관계를 파악하기 위해서 CCR 모형과 BCC 모형의 효율성 값과 entropy, bootstrap, super efficiency, cross efficiency 모형의 효율성 값들 간에 순위상관분석을 실시하였다. 본 연구를 통해 도출된 연구결과를 요약하면 다음과 같다. 첫째, 중요도와 변동계수를 이용한 모형들의 변별력 순위가 동일한 것으로 분석되어 변동계수를 DEA 모형의 변별력 평가기준으로 이용할 수 있다는 것이다. 둘째, 본 연구의 실증분석 결과에 따르면 4개 모형 중 super efficiency 모형이 변별력이 가장 높은 것으로 분석되었다. 셋째, CCR 모형과 순위상관관계가 가장 높은 모형은 super efficiency 모형으로 나타났고, BCC 모형과 순위상관관계가 가장 높은 모형도 super efficiency 모형으로 분석되었다.

• ETRI Journal
• /
• 제17권1호
• /
• pp.11-22
• /
• 1995

This paper proposes a modified error function to improve the error back-propagation (EBP) algorithm for multi-Layer perceptrons (MLPs) which suffers from slow learning speed. It can also suppress over-specialization for training patterns that occurs in an algorithm based on a cross-entropy cost function which markedly reduces learning time. In the similar way as the cross-entropy function, our new function accelerates the learning speed of the EBP algorithm by allowing the output node of the MLP to generate a strong error signal when the output node is far from the desired value. Moreover, it prevents the overspecialization of learning for training patterns by letting the output node, whose value is close to the desired value, generate a weak error signal. In a simulation study to classify handwritten digits in the CEDAR [1] database, the proposed method attained 100% correct classification for the training patterns after only 50 sweeps of learning, while the original EBP attained only 98.8% after 500 sweeps. Also, our method shows mean-squared error of 0.627 for the test patterns, which is superior to the error 0.667 in the cross-entropy method. These results demonstrate that our new method excels others in learning speed as well as in generalization.

• International Journal of Naval Architecture and Ocean Engineering
• /
• 제13권1호
• /
• pp.115-125
• /
• 2021

Simulation-based hull form optimization is a typical HEB (high-dimensional, expensive computationally, black-box) problem. Conventional optimization algorithms easily fall into the "curse of dimensionality" when dealing with HEB problems. A recently proposed Cross-Entropy (CE) optimization algorithm is an advanced stochastic optimization algorithm based on a probability model, which has the potential to deal with high-dimensional optimization problems. Currently, the CE algorithm is still in the theoretical research stage and rarely applied to actual engineering optimization. One reason is that the Monte Carlo (MC) method is used to estimate the high-dimensional integrals in parameter update, leading to a large sample size. This paper proposes an improved CE algorithm based on quasi-Monte Carlo (QMC) estimation using high-dimensional truncated Sobol subsequence, referred to as the QMC-CE algorithm. The optimization performance of the proposed algorithm is better than that of the original CE algorithm. With a set of identical control parameters, the tests on six standard test functions and a hull form optimization problem show that the proposed algorithm not only has faster convergence but can also apply to complex simulation optimization problems.

• Geomechanics and Engineering
• /
• 제19권3호
• /
• pp.283-293
• /
• 2019

Evaluating the stability of the excavation face of the cross-river shield tunnel with good accuracy is considered as a nonlinear and multivariable complex issue. Understanding the stability evaluation method of the shield tunnel excavation face is vital to operate and control the shield machine during shield tunneling. Considering the instability mechanism of the excavation face of the cross-river shield and the characteristics of this engineering, seven evaluation indexes of the stability of the excavation face were selected, i.e., the over-span ratio, buried depth of the tunnel, groundwater condition, soil permeability, internal friction angle, soil cohesion and advancing speed. The weight of each evaluation index was obtained by using the analytic hierarchy process and the entropy weight method. The evaluation model of the cross-river shield construction excavation face stability is established based on the idea point method. The feasibility of the evaluation model was verified by the engineering application in a cross-river shield tunnel project in China. Results obtained via the evaluation model are in good agreement with the actual construction situation. The proposed evaluation method is demonstrated as a promising and innovative method for the stability evaluation and safety construction of the cross-river shield tunnel engineerings.

• 응용통계연구
• /
• 제3권2호
• /
• pp.91-105
• /
• 1990

베이시안 결정론에서 사전 확률 분포함수는 표본을 추출하기 이전에 추정하여야 한다. 대개 는 분포함수군을 먼저 선택한 후, 그 중 하나를 결정자의 경험을 통하여 선택한다. 이러한 주관적인 사전 확률 분포함수의 선택방법이 베이시안 결정론에 대한 주요비판이 항상 되어 왔다. 본 논문에서는 최대 엔트로피 이론을 이용하여 우리 주변의 의사결정에 많이 이용되 는 정보들에 관한 객관적인 사전 확률 분포함수들을 구하였다. 그 결과는 히스토그램 형태 의 분포함수가 된다. 그러나 사전 정보가 많은 경우에는 최대 엔트로피 모형의 해를 구하기 위하여 복잡한 비선형 연립방정식을 풀어야 하는데, 구체적인 형태의 함수를 구하지 못하는 경우가 대부분이다. 이 때에는 초소의 크로스 엔트로피 모형을 이용하여 사전확률 분포함수 를 구하는 것이 편리하다. 그밖에 엔트로피 이론으로 구한 사전확률 분포함수의 확률적 수 렴성을 증명하였다.

• 한국산학기술학회논문지
• /
• 제9권1호
• /
• pp.154-159
• /
• 2008

본 연구에서는 수자원 기초자료 중 매우 중요하고 효과적인 유량측정방법을 위해 가장 먼저 결정하여야할 평균유속을 어떻게 결정할지를 제안코자 하며, 이를 위하여 현재까지 가장 널리 응용되고 있는 평균유속공식인 Manning공식과 최근에 그 효용성이 입증된 Chiu의 유속공식과의 상호관계에 대하여 분석 검토하여, 수로경사가 변화하는 경우나 유사유무에 관계없이 주어진 단면에서의 엔트로피 값, 즉 평형상태를 유지하려는 경향이 있음을 증명하였다. 따라서 인공수로에 관련된 간단한 수리입력 자료만 있다면 그동안 취득하기 어려운 $u_{max}$와 전체유속분포 산정에 매우 유용하게 사용될 수 있다고 사료된다.

• International Journal of Computer Science & Network Security
• /
• 제23권11호
• /
• pp.190-194
• /
• 2023

By looking the importance of communication, data delivery and access in various sectors including governmental, business and individual for any kind of data, it becomes mandatory to identify faults and flaws during cyber communication. To protect personal, governmental and business data from being misused from numerous advanced attacks, there is the need of cyber security. The information security provides massive protection to both the host machine as well as network. The learning methods are used for analyzing as well as preventing various attacks. Machine learning is one of the branch of Artificial Intelligence that plays a potential learning techniques to detect the cyber-attacks. In the proposed methodology, the Decision Tree (DT) which is also a kind of supervised learning model, is combined with the different cross-validation method to determine the accuracy and the execution time to identify the cyber-attacks from a very recent dataset of different network attack activities of network traffic in the UNSW-NB15 dataset. It is a hybrid method in which different types of attributes including Gini Index and Entropy of DT model has been implemented separately to identify the most accurate procedure to detect intrusion with respect to the execution time. The different DT methodologies including DT using Gini Index, DT using train-split method and DT using information entropy along with their respective subdivision such as using K-Fold validation, using Stratified K-Fold validation are implemented.