• Smart Structures and Systems
• /
• 제16권2호
• /
• pp.243-262
• /
• 2015

Optimal sensor placement (OSP) is a critical issue in construction and implementation of a sophisticated structural health monitoring (SHM) system. The uncertainties in the identified structural parameters based on the measured data may dramatically reduce the reliability of the condition evaluation results. In this paper, the information entropy, which provides an uncertainty metric for the identified structural parameters, is adopted as the performance measure for a sensor configuration, and the OSP problem is formulated as the multi-objective optimization problem of extracting the Pareto optimal sensor configurations that simultaneously minimize the appropriately defined information entropy indices. The nondirective movement glowworm swarm optimization (NMGSO) algorithm (based on the basic glowworm swarm optimization (GSO) algorithm) is proposed for identifying the effective Pareto optimal sensor configurations. The one-dimensional binary coding system is introduced to code the glowworms instead of the real vector coding method. The Hamming distance is employed to describe the divergence of different glowworms. The luciferin level of the glowworm is defined as a function of the rank value (RV) and the crowding distance (CD), which are deduced by non-dominated sorting. In addition, nondirective movement is developed to relocate the glowworms. A numerical simulation of a long-span suspension bridge is performed to demonstrate the effectiveness of the NMGSO algorithm. The results indicate that the NMGSO algorithm is capable of capturing the Pareto optimal sensor configurations with high accuracy and efficiency.

• 한국생물정보학회:학술대회논문집
• /
• 한국생물정보시스템생물학회 2005년도 BIOINFO 2005
• /
• pp.183-187
• /
• 2005

In this paper, we propose a heuristic method to select features using a Two-Phase Markov Blanket-based (TPMB) algorithm. The first phase, filtering phase, of TPMB algorithm works by filtering the obviously redundant features. A non-linear correlation method based on Information theory is used as a metric to measure the redundancy of a feature [1]. In second phase, approximating phase, the Markov Blanket (MB) of a system is estimated by employing the concept of cross entropy to identify the MB. We perform experiments on microarray data and report two popular dataset, AML-ALL [3] and colon tumor [4], in this paper. The experimental results show that the TPMB algorithm can significantly reduce the number of features while maintaining the accuracy of the classifiers.

• Journal of information and communication convergence engineering
• /
• 제13권2호
• /
• pp.113-122
• /
• 2015

Feature subset selection is as a pre-processing step in learning algorithms. In this paper, we propose an efficient algorithm, ModifiedFAST, for feature subset selection. This algorithm is suitable for text datasets, and uses the concept of information gain to remove irrelevant and redundant features. A new optimal value of the threshold for symmetric uncertainty, used to identify relevant features, is found. The thresholds used by previous feature selection algorithms such as FAST, Relief, and CFS were not optimal. It has been proven that the threshold value greatly affects the percentage of selected features and the classification accuracy. A new performance unified metric that combines accuracy and the number of features selected has been proposed and applied in the proposed algorithm. It was experimentally shown that the percentage of selected features obtained by the proposed algorithm was lower than that obtained using existing algorithms in most of the datasets. The effectiveness of our algorithm on the optimal threshold was statistically validated with other algorithms.

• 수산해양기술연구
• /
• 제49권1호
• /
• pp.18-28
• /
• 2013

It was compared the estimated parameters by the surplus production from three different models, i.e., three types (Schaefer, Gulland, and Schnute) of the traditional surplus production models, a stock production model incorporating covariates (ASPIC) model and a maximum entropy (ME) model. We also evaluated the performance of models in the estimation of their parameters. The maximum sustainable yield (MSY) of small yellow croaker (Pseudosciaena polyactis) in Korean waters ranged from 35,061 metric tons (mt) by Gulland model to 44,844mt by ME model, and fishing effort at MSY ($f_{MSY}$) ranged from 262,188hauls by Schnute model to 355,200hauls by ME model. The lowest root mean square error (RMSE) for small yellow croaker was obtained from the Gulland surplus production model, while the highest RMSE was from Schnute model. However, the highest coefficient of determination ($R^2$) was from the ME model, but the ASPIC model yielded the lowest coefficient. On the other hand, the MSY of Kapenta (Limnothrissa miodon) ranged from 16,880 mt by ASPIC model to 25,373mt by ME model, and $f_{MSY}$, from 94,580hauls by ASPIC model to 225,490hauls by Schnute model. In this case, both the lowest root mean square error (RMSE) and the highest coefficient of determination ($R^2$) were obtained from the ME model, which showed relatively better fits of data to the model, indicating that the ME model is statistically more stable and robust than other models. Moreover, the ME model could provide additional ecologically useful parameters such as, biomass at MSY ($B_{MSY}$), carrying capacity of the population (K), catchability coefficient (q) and the intrinsic rate of population growth (r).

• 정보처리학회논문지:컴퓨터 및 통신 시스템
• /
• 제13권1호
• /
• pp.31-37
• /
• 2024

기존의 비밀 공유방법들은 무작위적으로 생성된 다항식(polynomial) 또는 평면(plane)으로부터 지분을 유도하기 때문에 복잡하고 긴(complex and long) 형태의 지분을 생성한다. 그렇게 생성된 지분은 암기(not memorizable)가 불가능하고 관리가 어려우며, 이에 따라 지분을 보관하고 관리하기 위해 컴퓨터 시스템 또는 별도의 디지털 장치가 있어야 한다. 전통적인 비밀 공유방법을 변형하면 패스워드나 생체정보와 같이 사용자가 미리 선택한 값을 지분으로 설정하는 것이 가능할 수 있다. 그러나 이러한 방법은 사용자가 선택한 값의 무작위성(randomness) 또는 엔트로피(entropy)가 낮으면 완전 보안성을 보장하지 못할 수 있다. 즉, (t-1) 개 이하의 지분으로 비밀을 유추해내는 것이 가능해질 수 있다. 본 연구에서는 암기가 가능한 패스워드나 지문과 같은 생체정보와 같이 미리 지정된 값을 지분으로 사용하면서 비밀 공유의 보안성을 보장할 수 있는 지분 강화(share hardening) 방법을 제안한다.

• Investigative Magnetic Resonance Imaging
• /
• 제13권1호
• /
• pp.22-30
• /
• 2009

목적 : 자기공명영상화는 보통 긴 스캔 시간으로 인해 환자의 움직임이 큰 문제가 된다. 이러한 환자의 움직임을 보정하기 위한 한가지 방법인 영상의 엔트로피(entropy)를 이용한 후처리 방법은 다른 추가 데이터 획득 없이 효과적으로 움직임 인공물을 줄일 수 있음을 보였다. 하지만 이 방법의 가장 큰 문제는 처리 시간이 매우 길다는데 있다. 본 연구에서는 움직임 보정 처리시간을 줄이는 방법을 제안한다. 대상 및 방법 : 전체적인 보정 시간을 줄이기 위한 첫 번째 방법은, 퓨리에 변환의 분리성을 이용하여 전체적인 퓨리에 변환시간을 줄일 수 있다. 영상의 엔트로피 기준을 대신해 영상의 전체 화소의 합(pixel sum)을 움직임 보정의 기준으로 이용하여 영상 기준을 계산하는 시간을 절반 이하로 줄일 수 있다. 마지막으로 부분 퓨리에 재구성 방법을 조합하여 움직임의 영향을 보정하는 k-공간의 데이터 량을 줄임으로써 전체적인 처리 시간을 큰 폭으로 줄일 수 있다. 결과 : 제안한 방법을 사용하여 보정한 영상의 품질은 엔트로피를 이용한 보정 방법과 거의 흡사했으며, 대신 전체적인 처리 시간을 2차원 영상에서 15%로, 3차원 영상에서 30%로 줄일 수 있었다. 결론 : 제안한 방법을 병렬 영상화 기법 등과 결합하여 영상 보정 시간을 더욱 줄일 수 있을 것으로 기대한다. 제안하는 방법은 다른 보정 기법을 사용할 수 없을 때, 영상에서 움직임의 영향을 줄이는 방법으로 유용할 것으로 기대한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권5호
• /
• pp.1717-1732
• /
• 2015

The widespread use of location-based services (LBSs), which allows untrusted service provider to collect large number of user request records, leads to serious privacy concerns. In response to these issues, a number of LBS privacy protection mechanisms (LPPMs) have been recently proposed. However, the evaluation of these LPPMs usually disregards the background knowledge that the adversary may possess about users' contextual information, which runs the risk of wrongly evaluating users' query privacy. In this paper, we address these issues by proposing a generic formal quantification framework,which comprehensively contemplate the various elements that influence the query privacy of users and explicitly states the knowledge that an adversary might have in the context of query privacy. Moreover, a way to model the adversary's attack on query privacy is proposed, which allows us to show the insufficiency of the existing query privacy metrics, e.g., k-anonymity. Thus we propose two new metrics: entropy anonymity and mutual information anonymity. Lastly, we run a set of experiments on datasets generated by network based generator of moving objects proposed by Thomas Brinkhoff. The results show the effectiveness and efficient of our framework to measure the LPPM.

• Journal of information and communication convergence engineering
• /
• 제5권3호
• /
• pp.233-238
• /
• 2007

An approach to identify Chinese event types is proposed in this paper which combines a good feature selection policy and a Maximum Entropy (ME) model. The approach not only effectively alleviates the problem that classifier performs poorly on the small and difficult types, but improve overall performance. Experiments on the ACE2005 corpus show that performance is satisfying with the 83.5% macro - average F measure. The main characters and ideas of the approach are: (1) Optimal feature set is built for each type according to local feature selection, which fully ensures the performance of each type. (2) Positive and negative features are explicitly discriminated and combined by using one - sided metrics, which makes use of both features' advantages. (3) Wrapper methods are used to search new features and evaluate the various feature subsets to obtain the optimal feature subset.

• 품질경영학회지
• /
• 제47권3호
• /
• pp.537-552
• /
• 2019

Purpose: The development of information technology makes it easy to utilize high-dimensional categorical data. In this regard, the purpose of this study is to propose a novel method to select the proper categorical variables in high-dimensional categorical data. Methods: The proposed feature selection method consists of three steps: (1) The first step defines the goodness-to-pick measure. In this paper, a categorical variable is relevant if it has relationships among other variables. According to the above definition of relevant variables, the goodness-to-pick measure calculates the normalized conditional entropy with other variables. (2) The second step finds the relevant feature subset from the original variables set. This step decides whether a variable is relevant or not. (3) The third step eliminates redundancy variables from the relevant feature subset. Results: Our experimental results showed that the proposed feature selection method generally yielded better classification performance than without feature selection in high-dimensional categorical data, especially as the number of irrelevant categorical variables increase. Besides, as the number of irrelevant categorical variables that have imbalanced categorical values is increasing, the difference in accuracy between the proposed method and the existing methods being compared increases. Conclusion: According to experimental results, we confirmed that the proposed method makes it possible to consistently produce high classification accuracy rates in high-dimensional categorical data. Therefore, the proposed method is promising to be used effectively in high-dimensional situation.

• Smart Structures and Systems
• /
• 제31권4호
• /
• pp.335-349
• /
• 2023

Bridges constantly undergo deterioration and damage, the most common ones being concrete damage and exposed rebar. Periodic inspection of bridges to identify damages can aid in their quick remediation. Likewise, identifying components can provide context for damage assessment and help gauge a bridge's state of interaction with its surroundings. Current inspection techniques rely on manual site visits, which can be time-consuming and costly. More recently, robotic inspection assisted by autonomous data analytics based on Computer Vision (CV) and Artificial Intelligence (AI) has been viewed as a suitable alternative to manual inspection because of its efficiency and accuracy. To aid research in this avenue, this study performs a comparative assessment of different architectures, loss functions, and ensembling strategies for the autonomous segmentation of bridge components and damages. The experiments lead to several interesting discoveries. Nested Reg-UNet architecture is found to outperform five other state-of-the-art architectures in both damage and component segmentation tasks. The architecture is built by combining a Nested UNet style dense configuration with a pretrained RegNet encoder. In terms of the mean Intersection over Union (mIoU) metric, the Nested Reg-UNet architecture provides an improvement of 2.86% on the damage segmentation task and 1.66% on the component segmentation task compared to the state-of-the-art UNet architecture. Furthermore, it is demonstrated that incorporating the Lovasz-Softmax loss function to counter class imbalance can boost performance by 3.44% in the component segmentation task over the most employed alternative, weighted Cross Entropy (wCE). Finally, weighted softmax ensembling is found to be quite effective when used synchronously with the Nested Reg-UNet architecture by providing mIoU improvement of 0.74% in the component segmentation task and 1.14% in the damage segmentation task over a single-architecture baseline. Overall, the best mIoU of 92.50% for the component segmentation task and 84.19% for the damage segmentation task validate the feasibility of these techniques for autonomous bridge component and damage segmentation using RGB images.