• 한국구조물진단유지관리공학회 논문집
• /
• 제7권1호
• /
• pp.147-156
• /
• 2003

In this research, multiple corrosion factors of buried environments were measured in order to establish a formula for the corrosion character of corrugated steel structures in domestic environments. By substituting corrosion factors for each predicting formula, the durable lifetime was measured, and the measured lifetime was compared with the estimated lifetime by applying existing thickness-measuring techniques. A new usage standard was proposed with these results, in order to create the conclusion below. There are known differences in the soil factors used as variables in estimating the duration caused by the seasonal effects of rainfall and temperature. Comparing the durable lifetime estimated by each predicting formula, the findings show that the California technique is conservative. This study demonstrates that the error range of the AISI technique, which is mostly used as a duration technique, is a very narrow predicting technique as compared with many other countries. Considering that there is on average, a 13% error margin in this study, a proposed safety factor of 0.87 could be used to more accurately predict the duration. The laying time in the California technique is not longer than the whole durability, and as a result, this error margin exists. It is concluded that this study on the open area has been overdue. Based on these findings, it's proposed that this error margin should be applied to the domestic environment through periodic observation, in order to establish the predicting techniques of durable lifetime.

• Journal of Information Technology Applications and Management
• /
• 제13권4호
• /
• pp.273-290
• /
• 2006

This study suggests a frequency matrix technique to predict personal credit rate more efficiently using incomplete data sets. At first this study test on multiple discriminant analysis and logistic regression analysis for predicting personal credit rate with incomplete data sets. Missing values are predicted with mean imputation method and regression imputation method here. An artificial neural network and frequency matrix technique are also tested on their performance in predicting personal credit rating. A data set of 8,234 customers in 2004 on personal credit information of Bank A are collected for the test. The performance of frequency matrix technique is compared with that of other methods. The results from the experiments show that the performance of frequency matrix technique is superior to that of all other models such as MDA-mean, Logit-mean, MDA-regression, Logit-regression, and artificial neural networks.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2005년도 Proceedings of ISRS 2005
• /
• pp.186-188
• /
• 2005

Because forest fire changes the direction according to the environmental elements, it is difficult to predict the direction of it. Currently, though some researchers have been studied to which predict the forest fire occurrence and the direction of it, using the remote detection technique, it is not enough and efficient. And recently because of the development of the sensor technique, a lot of In-Situ sensors are being developed. These kinds of In-Situ sensor data are used to collect the environmental elements such as temperature, humidity, and the velocity of the wind. Accordingly we need the prediction technique about the environmental elements analysis and the direction of the forest fire, using the In-Situ sensor data. In this paper, as a technique for predicting the direction of the forest fire, we propose the correlation analysis technique about In-Situ sensor data such as temperature, humidity, the velocity of the wind. The proposed technique is based on the clustering method and clusters the In-Situ sensor data. And then it analyzes the correlation of the multivariate correlations among clusters. These kinds of prediction information not only helps to predict the direction of the forest fire, but also finds the solution after predicting the environmental elements of the forest fire. Accordingly, this technique is expected to reduce the damage by the forest fire which occurs frequently these days.

• 교육녹색환경연구
• /
• 제11권3호
• /
• pp.46-54
• /
• 2012

The accuracy of cost estimation at an early stage in school building project is one of the critical factors for successful completion. So various of techniques are developed to predict the construction cost accurately and expeditely. Among the techniques, Support Vector Machine(SVM) has an excellent ability for generalization performance. Therefore, the purpose of this study is to construct the prediction model for construction cost of educational building project using support vector machine technique. And to verify the accuracy of prediction model for construction cost. The performance data used in this study are 217 school building project cost which have been completed from 2004 to 2007 in Gyeonggi-Do, Korea. The result shows that average error rate was 7.48% for SVM prediction model. So using SVM model on predicting construction cost of educational building project will be a considerably effective way at the early project stage.

• Journal of Communications and Networks
• /
• 제17권6호
• /
• pp.568-581
• /
• 2015

Predicting locations of users with portable devices such as IP phones, smart-phones, iPads and iPods in public wireless local area networks (WLANs) plays a crucial role in location management and network resource allocation. Many techniques in machine learning and data mining, such as sequential pattern mining and clustering, have been widely used. However, these approaches have two deficiencies. First, because they are based on profiles of individual mobility behaviors, a sequential pattern technique may fail to predict new users or users with movement on novel paths. Second, using similar mobility behaviors in a cluster for predicting the movement of users may cause significant degradation in accuracy owing to indistinguishable regular movement and random movement. In this paper, we propose a novel fusion technique that utilizes mobility rules discovered from multiple similar users by combining clustering and sequential pattern mining. The proposed technique with two algorithms, named the clustering-based-sequential-pattern-mining (CSPM) and sequential-pattern-mining-based-clustering (SPMC), can deal with the lack of information in a personal profile and avoid some noise due to random movements by users. Experimental results show that our approach outperforms existing approaches in terms of efficiency and prediction accuracy.

• Safety and Health at Work
• /
• 제15권2호
• /
• pp.220-227
• /
• 2024

Background: Though the artificial neural network (ANN) technique has been used to predict noise-induced hearing loss (NIHL), the established prediction models have primarily relied on cross-sectional datasets, and hence, they may not comprehensively capture the chronic nature of NIHL as a disease linked to long-term noise exposure among workers. Methods: A comprehensive dataset was utilized, encompassing eight-year longitudinal personal hearing threshold levels (HTLs) as well as information on seven personal variables and two environmental variables to establish NIHL predicting models through the ANN technique. Three subdatasets were extracted from the afirementioned comprehensive dataset to assess the advantages of the present study in NIHL predictions. Results: The dataset was gathered from 170 workers employed in a steel-making industry, with a median cumulative noise exposure and HTL of 88.40 dBA-year and 19.58 dB, respectively. Utilizing the longitudinal dataset demonstrated superior prediction capabilities compared to cross-sectional datasets. Incorporating the more comprehensive dataset led to improved NIHL predictions, particularly when considering variables such as noise pattern and use of personal protective equipment. Despite fluctuations observed in the measured HTLs, the ANN predicting models consistently revealed a discernible trend. Conclusions: A consistent correlation was observed between the measured HTLs and the results obtained from the predicting models. However, it is essential to exercise caution when utilizing the model-predicted NIHLs for individual workers due to inherent personal fluctuations in HTLs. Nonetheless, these ANN models can serve as a valuable reference for the industry in effectively managing its hearing conservation program.

• 한국지구물리탐사학회:학술대회논문집
• /
• 한국지구물리탐사학회 2003년도 Proceedings of the international symposium on the fusion technology
• /
• pp.129-141
• /
• 2003

The precise prediction of ground displacement plays an important role in planning and constructing tunnels. In this study, an equation for predicting the surface and crown settlement is suggested by examining the theories of ground movement caused by tunnel excavation. From the 3D numerical modeling, the reinforcement effect of UAM (Umbrella Arch Method) is quantitatively analyzed with respect to deformation modulus and overburden. By using a regression technique for the numerical results, an equation for predicting the settlement is suggested.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2008년도 춘계학술대회 논문집
• /
• pp.109-117
• /
• 2008

In this study, a method predicting the displacement responseof structures from the measured dynamic strain signal is proposed by using a mode decomposition technique. Dynamic loadings including wind and seismic loadings could be exerted to the bridge. In order to examine the bridge stability against these dynamic loadings, the prediction of displacement response is very important to evaluate bridge stability. Because it may be not easy for the displacement response to be acquired directly on site, an indirect method to predict the displacement response is needed. Thus, as an alternative for predicting the displacement response indirectly, the conversion of the measured strain signal into the displacement response is suggested, while the measured strain signal can be obtained using fiber optic Bragg-grating (FBG) sensors. To overcome such a problem, a mode decomposition technique was used in this study. The measured strain signal is decomposed into each modal component by using the empirical mode decomposition(EMD) as one of mode decomposition techniques. Then, the decomposed strain signals on each modal component are transformed into the modal displacement components. And the corresponding mode shapes can be also estimated by using the proper orthogonal decomposition(POD) from the measured strain signal. Thus, total displacement response could be predicted from combining the modal displacement components.

• 대한토목학회논문집
• /
• 제28권4A호
• /
• pp.507-515
• /
• 2008

본 연구에서는 모드분해기법을 이용한 변형률신호로부터 변위응답추정 방법을 개발하였다. 일반적으로 교량의 안정성평가는 완공 후에 초점이 맞추어져 있다. 하지만 가설 중에도 풍하중과 지진하중과 같은 동적하중에 노출되어 있으며, 이런 동적하중에 대한 안정성을 검토하기 위해 교량의 안정성 평가에 있어 중요한 인자인 변위를 추정하는 것이 중요하다. 그러나 건설현장에서의 적절한 변위측정 방법의 부재로 인하여 대형구조물의 전체적인 변위를 측정할 수 없는 것이 현실이다. 본 연구에서는 간접적으로 변위를 추정하는 방법인 변형률로 변위를 추정하는 방법을 제시하였으며, 광섬유 브래그 격자 센서(fiber optic Bragg-grating sensor)를 사용하여 변형률을 계측하였다. 기존에도 FBG센서를 이용한 변위추정 방법이 있었으며 기존의 방법으로는 정적하중에 대한 변위추정은 가능하였으나 고차 모드의 변형률신호와 노이즈의 영향 때문에 동적하중에 대한 변위추정은 많은 오차가 발생하여 정확한 변위추정이 어려웠다. 이런 오차를 줄이는 방법으로 모드분해기법을 사용하였다. 모드분해기법은 변형률신호로부터 proper orthogonal decomposition(POD)을 이용하여 추정한 모드형상과 empirical mode decomposition(EMD)을 이용하여 모드 분해한 변형률신호로 모드별 변위응답을 추정하고, 구조물의 주요 모드에 대한 변위응답을 합하여 전체변위응답을 추정하는 방법이다. 제안한 모드분해기법을 검증하기 위해 실내모형실험을 수행하였다.

• 한국근적외분광분석학회:학술대회논문집
• /
• 한국근적외분광분석학회 2001년도 NIR-2001
• /
• pp.1618-1618
• /
• 2001

Predicting quality traits using near infrared (NIR) spectroscopy on whole grain samples has gained wide acceptance as a non-destructive, rapid and cost effective technique. Barley breeding programs throughout southern Australia currently use this technology as a tool for selecting malting quality lines. For the past 3 years whole grain barley calibrations have been developed at VIDA to predict malting quality traits in the early generation selections of the breeding program. More recently calibrations for whole grain malt have been developed and introduced to aid in selecting malted samples at the mid-generation stage for more complex malting quality traits. Using the same population set, barley and malt calibrations were developed to predict hot water extracts (EBC and IoB), diastatic power, free $\alpha$-amino nitrogen, soluble protein, wort $\beta$-glucan and $\beta$-glucanase. The correlation coefficients between NIR predicted values and laboratory methods for malt were all highly significant ($R^2$ > 0.84), whereas the correlation coefficients for the barley calibrations were lower ($R^2$ > 0.57) but still significant. The magnitude of the error in predicting hot water extract, diastatic power and wort $\beta$-glucan using whole grain malt was reduced by 50% when compared with predicting the same trait using whole grain barley. This can be explained by the complex nature of attempting to develop calibrations on whole grain barley utilizing malt data. During malting, the composition of barley is modified by the action of enzymes throughout the steeping and germination stages and by heating during the kilning stage. Predicting malting quality on whole grain malt is a more reliable alternative to predicting whole grain barley, although there is the added expense of micro-malting the samples. The ability to apply barley and malt calibrations to different generations is an advantage to a barley breeding program that requires thousands of samples to be assessed each year.