• Journal of the Earthquake Engineering Society of Korea
• /
• v.10 no.4 s.50
• /
• pp.1-13
• /
• 2006

Soil and rock dynamic properties such as shear wave velocity $(V_s)$, compressional wave velocity $(V_p)$ and corresponding Poisson's ratio (v) are very important geotechnical parameters in predicting deformational behavior of structures as well as practicing seismic design and performance evaluation. In an effort to measure the parameter efficiently and accurately, various bore-hole seismic testing techniques have been, thus, developed and used during past several decades. In this study, cross-hole seismic testing technique which is known as the most reliable seismic method was adopted for obtaining geotechnical dynamic properties. To perform successfully the cross-hole test for rock as well as soil layers regardless of the ground water level, spring-loaded source which impact laterally a subsurface ground in vertical bore-hole was developed and applied at three study areas, which contain four sites composed of two existing port sites and two new LNG storage facility sites. The geotechnical dynamic properties such as $V_s,\;V_p$ and v with depth from the soil surface to the engineering and seismic bedrock were efficiently determined from the laterally impacted cross-hole seismic tests at study sites, and were provided as the fundamental parameters for the seismic performance evaluation of the existing ports and the seismic design of the LNG storage facilities.

• Journal of Genetic Medicine
• /
• v.8 no.1
• /
• pp.28-34
• /
• 2011

Most clinicians understand clinical trials as the evaluation process for new medicine before their use. However, clinical trials can also be applied to laboratory diagnostic tests (LDTs) to verify diagnostic accuracy and efficacy before their clinical laboratory implementation for patients. The clinical trial of LDT has two distinctive characteristics that are different from the case of pharmaceuticals and thus worth special consideration. One of them is the level of evidence. The well-designed randomized controlled trials (RCTs) are known to provide the best evidence to prove the clinical efficacy of any pharmaceutical products. However, RCTs lose practicality when applied to LDTs due to various issues including ethical complications. For this reason, comparative study format is considered more feasible approach for LDTs. In addition pharmaceuticals and LDTs are different in that the user's intervention is not required for the former but critical to the latter. Moreover, in the case of pharmaceuticals, end-products are produced by manufacturers before being used by clinicians. However, in LDTs, once reagents and instruments are provided by manufacturers, they are first utilized by clinical laboratories to produce test results in order for clinicians to use them later. In other words, when it comes to LDTs, clinical laboratories play the role of manufacturers, providing reliable test results with improved quality assurance. Considering the distinctive characteristics of LDTs, we would like to offer detailed suggestions to successfully perform clinical trials in LDTs, which include analytical performance measures, clinical test performance measures, diagnostic test accuracy measures, clinical effectiveness measures, and post-implementation surveillance.

• Journal of Wetlands Research
• /
• v.25 no.3
• /
• pp.167-175
• /
• 2023

Precipitation data is one of the essential input datasets used in various fields such as wetland management, hydrological simulation, and water resource management. In order to efficiently manage water resources using precipitation data, it is essential to secure as much data as possible by minimizing the missing rate of data. In addition, more efficient hydrological simulation is possible if precipitation data for ungauged areas are secured. However, missing precipitation data have been estimated mainly by statistical equations. The purpose of this study is to propose a new method to restore missing precipitation data using machine learning algorithms that can predict new data based on correlations between data. Moreover, compared to existing statistical methods, the applicability of machine learning techniques for restoring missing precipitation data is evaluated. Representative machine learning algorithms, Artificial Neural Network (ANN) and Random Forest (RF), were applied. For the performance of classifying the occurrence of precipitation, the RF algorithm has higher accuracy in classifying the occurrence of precipitation than the ANN algorithm. The F1-score and Accuracy values, which are evaluation indicators of the classification model, were calculated as 0.80 and 0.77, while the ANN was calculated as 0.76 and 0.71. In addition, the performance of estimating precipitation also showed higher accuracy in RF than in ANN algorithm. The RMSE of the RF and ANN algorithms was 2.8 mm/day and 2.9 mm/day, and the values were calculated as 0.68 and 0.73.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.4
• /
• pp.687-692
• /
• 2023

We analyze the accuracy of emotion analysis of natural language processing deep learning model and propose to use it for emotional content development. After looking at the outline of the GPT-3 model, about 6,000 pieces of dialogue data provided by Aihub were input to 9 emotion categories: 'joy', 'sadness', 'fear', 'anger', 'disgust', and 'surprise'. ', 'interest', 'boredom', and 'pain'. Performance evaluation was conducted using the evaluation indices of accuracy, precision, recall, and F1-score, which are evaluation methods for natural language processing models. As a result of the emotion analysis, the accuracy was over 91%, and in the case of precision, 'fear' and 'pain' showed low values. In the case of reproducibility, a low value was shown in negative emotions, and in the case of 'disgust' in particular, an error appeared due to the lack of data. In the case of previous studies, emotion analysis was mainly used only for polarity analysis divided into positive, negative, and neutral, and there was a limitation in that it was used only in the feedback stage due to its nature. We expand emotion analysis into 9 categories and suggest its use in the development of emotional content considering it from the planning stage. It is expected that more accurate results can be obtained if emotion analysis is performed by additionally collecting more diverse daily conversations through follow-up research.

• Journal of Korean Society of Transportation
• /
• v.36 no.1
• /
• pp.38-50
• /
• 2018

Emerging automated driving environments will lead to a mixed traffic flow depending on the interaction between automated vehicles (AVs) and manually driven vehicles (MVs) because the market penetration rate (MPR) of AVs will gradually increase over time. Understanding the characteristics of mixed traffic conditions, and developing a method to control both AV and MV maneuverings smoothly is a backbone of the traffic management in the era of automated driving. To facilitate smooth vehicle interactions, the maneuvering of AVs should be properly determined by various traffic and road conditions, which motivates this study. This study investigated whether the aggressiveness of AV maneuvering, defined as automated driving aggressiveness (ADA), affect the performance of mixed traffic flow. VISSIM microscopic simulation experiments were conducted to derive proper ADAs for satisfying both the traffic safety and the operational efficiency. Traffic conflict rates and average travel speeds were used as indicators for the performance of safety and operations. While conducting simulations, level of service(LOS) and market penetration rate(MPR) of AVs were also taken into considerations. Results implies that an effective guideline to manage the ADA under various traffic and road conditions needs to be developed from the perspective of traffic operations to optimize traffic performances.

• Journal of Wetlands Research
• /
• v.19 no.1
• /
• pp.52-62
• /
• 2017

In this study, land surface model was used to simulate the soil moisture of South and North Korea for the past 30 years, and the difference in their variation was analyzed. In addition, satellite observed soil moisture data provided by Soil Moisture CCI was analyzed to evaluate the simulation results of VIC model. For the comparison between the simulated and observed data, the CSEOF analysis was applied to indirectly assess the performance of the VIC model rather than simply comparing soil moisture values. The results of this study are summarized as follows. First, the annual variability of soil moisture showed a similar tendency in both South and North Korea, but it was found that the soil moisture in South Korea was as high as 1%, up to 7%, higher than the soil moisture in North Korea. Secondly, the soil moisture in spring between April to June is similar in South and North Korea, whereas the soil moisture after the rainy season is up to 40% in South Korea, but remains at maximum 32% in North Korea. Third, the overall simulated soil moisture is about 4% smaller than the satellite observed soil moisture, but the degree of increase over the past 30 years is similar to that of satellite observed soil moisture. Finally, a comparison of the CSEOF from the satellite observed soil moisture and the VIC model derived soil moisture showed that the soil moisture from April to June shows a much different pattern from each other. However, in July and October, there was a slight similarity, and it was confirmed that August and September has quite similar patterns.

• Journal of Internet Computing and Services
• /
• v.22 no.2
• /
• pp.29-39
• /
• 2021

As damages to individuals, private sectors, and businesses increase due to newly occurring cyber attacks, the underlying network security problem has emerged as a major problem in computer systems. Therefore, NIDS using machine learning and deep learning is being studied to improve the limitations that occur in the existing Network Intrusion Detection System. In this study, a deep learning-based NIDS model study is conducted using the Convolution Neural Network (CNN) algorithm. For the image classification-based CNN algorithm learning, a discrete algorithm for continuity variables was added in the preprocessing stage used previously, and the predicted variables were expressed in a linear relationship and converted into easy-to-interpret data. Finally, the network packet processed through the above process is mapped to a square matrix structure and converted into a pixel image. For the performance evaluation of the proposed model, NSL-KDD, a representative network packet data, was used, and accuracy, precision, recall, and f1-score were used as performance indicators. As a result of the experiment, the proposed model showed the highest performance with an accuracy of 85%, and the harmonic mean (F1-Score) of the R2L class with a small number of training samples was 71%, showing very good performance compared to other models.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.6
• /
• pp.272-277
• /
• 2019

In this paper, a manufacturing method of an anti-glare cover glass on LCD for outdoor use is proposed. The main specification of cover glass is hardness and anti-glare. Hardness is achieved by using the tempered glass, and anti-glare(AG) film is laminated to meet anti-glare specification no the tempered glass. However, the AG film is difficult to maintain the AG performance continuously because the abrasion resistance of the PET film itself is as weak as about 3H. Therefore, a novel production procedure using screen printing method is proposed. The proposed coating is implemented by applying $ZnO-B_2O_3-SiO_2$ powder on glass surface and the glass is made with enhanced hardness through tempering process. In order to apply the ZBS powder uniformly on the glass surface, a screen printing process is used. The main parameters to be considered in screen printing are the oil concentration and mesh opening size. Because the amount of ZBS powder applied to the printing process is controlled by these two parameters, the correlativity is confirmed through the experiments. In order to evaluate the performance of the proposed method, the haze, surface roughness and transmittance are selected as the performance index and are compared with the AG film. As a result of comparison, it is verified that the transmittance of the proposed tempered glass is 83.1%, which is slightly lower than 89.5% of AG film, but the hardness is more than double to 7H.

• The Journal of Sustainable Design and Educational Environment Research
• /
• v.16 no.2
• /
• pp.44-53
• /
• 2017

Facility management is to maintain and develop the primary structural, functional, aesthetic performance of facility in order to guaranty the users' daily convenience and safety. However it is hard to maintain and serve their intended function and safe environment from the beginning as times go by. As present educational government of city and local area has been performing formally facility check and management as well as maintenance of school facility, it is hard to respond a dangerous situation at the suitable time and safety prevention plans are delayed. In addition, educational environment improving budget have been unreasonably decided not according to the allocating criteria. Therefore, this research developed a same, simple, and quantitative evaluation method of deterioration level of elementary, middle, and high School in Korea and verified usability of the method through the case study.

• Journal of KIISE:Software and Applications
• /
• v.36 no.10
• /
• pp.861-873
• /
• 2009

There are many reasons that fail to get appropriate information in information retrieval. Allomorph is one of the reasons for search failure due to keyword mismatch. This research proposes a method to construct alternative word candidates automatically in order to minimize search failure due to keyword mismatch. Assuming that two words have similar meaning if they have similar co-occurrence words, the proposed method uses the concept of concentration, association word set, cosine similarity between association word sets and a filtering technique using confidence. Performance of the proposed method is evaluated using a manually extracted alternative list. Evaluation results show that the proposed method outperforms the context window overlapping in precision and recall.