• Journal of Internet Computing and Services
• /
• v.17 no.2
• /
• pp.49-57
• /
• 2016

Ensemble classification involves combining multiple classifiers to obtain more accurate predictions than those obtained using individual models. Ensemble learning techniques are known to be very useful for improving prediction accuracy. Bagging is one of the most popular ensemble learning techniques. Bagging has been known to be successful in increasing the accuracy of prediction of the individual classifiers. Bagging draws bootstrap samples from the training sample, applies the classifier to each bootstrap sample, and then combines the predictions of these classifiers to get the final classification result. Bootstrap samples are simple random samples selected from the original training data, so not all bootstrap samples are equally informative, due to the randomness. In this study, we proposed a new method for improving the performance of the standard bagging ensemble by optimizing bootstrap samples. A genetic algorithm is used to optimize bootstrap samples of the ensemble for improving prediction accuracy of the ensemble model. The proposed model is applied to a bankruptcy prediction problem using a real dataset from Korean companies. The experimental results showed the effectiveness of the proposed model.

• Korean Journal of Environmental Biology
• /
• v.36 no.3
• /
• pp.345-351
• /
• 2018

An efficient, quick and low-cost extraction and clean up method for the determination of 14 polycyclic aromatic hydrocarbons (PAHs) in the agricultural water samples was optimized using gas chromatography-tandem mass spectrometry (GC-MS/MS). The extraction of the target compounds in water sample was carried out with acetonitrile, followed by partitioning promoted by the addition of salt. As a clean-up procedure, dispersive solid phase extraction was employed to purify the analytes of interest for GC-MS/MS analysis. This method was successfully applied for the quantification of PAHs in real water samples collected for the purpose of monitoring from the waterways located in Chungbuk (15 sites) and Gyeongbuk (6 sites), S. Korea. Phenanthrene (0.54 to $2.53{\mu}gL^{-1}$) was detected in all the water samples collected from both the sites. Fluoranthene was detected in the water samples collected from the two sites in Gyeongbuk province, but other PAHs were not determined in these water sampling sites. Based on these results, the determined PAHs were conducted using an environmental risk assessment. The risk characterization ratios (RCRs) for phenanthrene ranged from 0.37 to 3.21. These RCR values referred to as risk was not controlled because RCR values of some sites were greater than 1. In conclusion, it is proposed that the optimized method in combination with GC-MS/MS could be successfully employed for the determination of PAHs in any environmental samples including water samples.

• Parasites, Hosts and Diseases
• /
• v.61 no.1
• /
• pp.15-23
• /
• 2023

Concerns about foodborne illnesses caused by Kudoa septempunctata are steadily growing, but reports of K. septempunctata in clinical and food specimens related to food poisoning in Korea are limited. This study aimed to genetically identify K. septempunctata in patients with acute diarrhea and in clinical and food samples related to food poisoning caused by sashimi consumption. Both real-time and nested polymerase chain reaction assays were performed to detect K. septempunctata 18S and 28S rDNA genes in the stools of 348 patients with acute diarrhea, 11 samples (6 stool and 5 rectal swab samples) from patients with food poisoning, and 2 raw Paralichthys olivaceus samples collected from a restaurant where a food poisoning incident occurred. K. septempunctata was identified in 5 clinical specimens (4 stools and 1 rectal swab) and 1 P. olivaceus sashimi sample. All detected K. septempunctata were of genotype ST3. This is the first study to identify K. septempunctata in both patients and food samples with epidemiological relevance in Korea, providing evidence that it is a pathogen that causes food poisoning. Also, this is the first study to confirm the presence of K. septempunctata genes in rectal swabs. Despite continuing suspected occurrences of Kudoa foodborne outbreaks, the rate of identification of K. septempunctata is very low. One reason for this is the limitation in obtaining stool and vomit samples for the diagnosis of Kudoa infection. We strongly suggest the inclusion of rectal swabs among the diagnostic specimens for Kudoa food poisoning.

• Fashion & Textile Research Journal
• /
• v.23 no.2
• /
• pp.247-260
• /
• 2021

While it is an important element of clothing construction, research has so far been very limited on the similarities between virtual and real clothing in terms of the type of neckline. The purpose of this study is to verify the similarity, accuracy of virtualization, and actuality of neckline, which all play an important role in individual impressions and image formation, and require considerable modification when fitting real samples. A total of 5 neckline models were selected through the analysis of dress composition textbooks. The selected designs were then planned and manufactured in muslin. The specimen clothes were then tested on a female model in her 20s. 2 kinds of virtual bodies were created in order to compare the real and the virtual dressing. The first virtual body was made through an Artec 3D Eva scan of the model, and the other was made by entering the model's measurements in a CLO 3D program. A visual image of the front, side, and back image of both the real and virtual dressing were subsequently collected. The collected images were then evaluated by 20 professional fashion workers who checked the similarity between the real and the virtual versions. The current study found that the similarity between the actual and virtual wearing of the five neckline designs with reality appeared higher with the virtual wearing image using the 3D-scanned body. The results of this study could provide further information on the selection of appropriate avatars to clothing companies that check the fit of clothing by utilizing 3D virtualized programs.

• Fisheries and Aquatic Sciences
• /
• v.24 no.11
• /
• pp.351-359
• /
• 2021

The red sea bream iridovirus (RSIV) belonging to genus Megalocytivirus is responsible for red sea bream iridoviral disease (RSIVD) in marine and freshwater fishes. Although several diagnostic assays for RSIV have been developed, diagnostic sensitivity (DSe) and specificity (DSp) of real-time polymerase chain reaction (PCR) assays are not yet evaluated. In this study, we developed a TaqMan probe-based real-time PCR method and evaluated its DSe and DSp. To detect RSIV, the probe and primers were designed based on consensus sequences of the major capsid protein (MCP) genes from megalocytiviruses including RSIV, infectious spleen and kidney necrosis virus (ISKNV), and turbot reddish body iridovirus (TRBIV). The probe and primers were shown to be specific for RSIV, ISKNV, and TRBIV-types megalocytiviruses. A 95% limit of detection (LOD_95%) was determined to be 5.3 viral genome copies/µL of plasmid DNA containing the MCP gene from RSIV. The DSe and DSp of the developed real-time PCR assay for field samples (n = 112) were compared with those of conventional PCR assays and found to be 100% and 95.2%, respectively. The quantitative results for SYBR Green and TaqMan probe-based real-time PCR were not significantly different. The TaqMan probe-based real-time PCR assay for RSIV may be used as an appropriate diagnostic tool for qualitative and quantitative analysis.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.3
• /
• pp.151-158
• /
• 2024

Although on-device artificial intelligence (AI) has gained attention to diagnosing machine faults in real time, most previous studies did not consider the model retraining and redeployment processes that must be performed in real-world industrial environments. Our study addresses this challenge by proposing an on-device AI-based real-time machine fault diagnosis system that utilizes continual learning. Our proposed system includes a lightweight convolutional neural network (CNN) model, a continual learning algorithm, and a real-time monitoring service. First, we developed a lightweight 1D CNN model to reduce the cost of model deployment and enable real-time inference on the target edge device with limited computing resources. We then compared the performance of five continual learning algorithms with three public bearing fault datasets and selected the most effective algorithm for our system. Finally, we implemented a real-time monitoring service using an open-source data visualization framework. In the performance comparison results between continual learning algorithms, we found that the replay-based algorithms outperformed the regularization-based algorithms, and the experience replay (ER) algorithm had the best diagnostic accuracy. We further tuned the number and length of data samples used for a memory buffer of the ER algorithm to maximize its performance. We confirmed that the performance of the ER algorithm becomes higher when a longer data length is used. Consequently, the proposed system showed an accuracy of 98.7%, while only 16.5% of the previous data was stored in memory buffer. Our lightweight CNN model was also able to diagnose a fault type of one data sample within 3.76 ms on the Raspberry Pi 4B device.

• Journal of Korean Society of Environmental Engineers
• /
• v.32 no.2
• /
• pp.121-130
• /
• 2010

Metals and sulfate can be considerably dissolved at low pH condition in the acid mine drainage(AMD) and it would make an environmental problems. There are only few of acid mine drainage treatment systems in Korea which are operating, but these still have an effect on the surrounding stream. In this study, quantification of indicator microorganisms was conducted to judge the environmental impact of AMD on microflora by quantitative real-time PCR in the drainage samples of four mines and the water samples of each surrounding stream. Two species of iron reducing bacteria(Rhodoferax ferrireducens T118 and Acidiphilium cryptum JF-5) were selected for indicator bacteria based on 16S rRNA cloning analysis, and sulfate reducing bacteria(Desulfosporosinus orientus), iron and sulfur oxidizing bacteria(Acidothiobacillus ferrooxidans) and iron oxidizing bacteria(Leptosprillum ferrooxidans) were included into indicator since these were found in the previous studies on the mining area. Thereafter, the comparative analysis of four mines were established by the microbiological variation index and it was determined that the biological environment effect of AMD is highest in Samtan mine which doesn t contain treatment system by the value.

• Journal of the Korean Chemical Society
• /
• v.56 no.1
• /
• pp.58-61
• /
• 2012

The solvent extraction of platinum (IV) metal from hydrochloric acid media using 4-(4-ethoxybenzylideneamino)-5-methyl-4H-1,2,4-triazole-3-thiol (EBIMTT) in chloroform was studied as a function of several variables, such as reagent, acid and metal ion concentration, effect of various diluents, and diverse ions. The proposed method was further applied for the separation of platinum (IV) from binary mixtures, synthetic mixtures, alloys and commercially available samples.

• International Journal of Quality Innovation
• /
• v.7 no.3
• /
• pp.58-69
• /
• 2006

This paper presents neural network-based recognition system for automatic light emitting diode (LED) inspection. The back-propagation neural network (BPNN) is proposed and tested. The current-voltage (I-V) characteristic data of LED from the inspection process is used for the network training and testing. This study selects 300 random samples as network training and employs 100 samples as network testing. The experimental results show that if the classification work is done well, the accuracy of recognition is 100%, and the testing speed of the proposed recognition system is almost one half faster than the traditional inspection system does. The proposed neural-network approach is successfully demonstrated by real data sets and can be effectively developed as a recognition system for a practical application purpose.

• Journal of the Korean Magnetic Resonance Society
• /
• v.13 no.1
• /
• pp.7-14
• /
• 2009

The spectral data were collected from the two 600 MHz spectrometers from the two major manufacturers, Broker and Varian. The samples were prepared to create standard curves for quantitative measurements of metabolite concentrations. Instead of employing one-dimensional $^1H$ experiments, the two-dimensional $^1H-^{13}C$ HSQC experiments were performed for better separation of resonances. For some resonances, the high salt condition hindered the linear correlation between the intensity and actual metabolite concentration. Excluding overlapped ones, most resonances showed good linearity. Although the Varian spectrometer showed better linearity, both spectrometers were able to generate acceptable standard curves. From this data, we could identify resonances that could be used to better quantify the concentrations of the particular metabolites. With these standard curves, the quantitative measurements of the metabolites from the real samples will be facilitated.