• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.2
• /
• pp.750-762
• /
• 2019

The importance of the call center as a contact point for the enterprise is growing. However, call centers have difficulty with their operating agents due to the agents' lack of knowledge and owing to frequent agent turnover due to downturns in the business, which causes deterioration in the quality of customer service. Therefore, through an N-bank call center case study, we developed a system to reduce the burden of keeping up business knowledge and to improve customer service quality. It is a "real-time agent advisor" system that provides agents with answers to customer questions in real time by combining AI technology for speech recognition, natural language processing, and questions & answers for existing call center information systems, such as a private branch exchange (PBX) and computer telephony integration (CTI). As a result of the case study, we confirmed that the speech recognition system for real-time call analysis and the corpus construction method improves the natural speech processing performance of the query response system. Especially with name entity recognition (NER), the accuracy of the corpus learning improved by 31%. Also, after applying the agent advisor system, the positive feedback rate of agents about the answers from the agent advisor was 93.1%, which proved the system is helpful to the agents.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.3
• /
• pp.16-22
• /
• 2019

Off-line programming and robot calibration through simulation are essential when setting up a robot in a robot automation production line. In this study, we developed a new robot calibration method to match the CAD data of the production line with the measurement data on the site using 3D scanner. The proposed method calibrates the robot using 3D point cloud data through Iterative Closest Point algorithm. Registration is performed in three steps. First, vertices connected by three planes are extracted from CAD data as feature points for registration. Three planes are reconstructed from the scan point data located around the extracted feature points to generate corresponding feature points. Finally, the transformation matrix is calculated by minimizing the distance between the feature points extracted through the ICP algorithm. As a result of applying the software to the automobile welding robot installation, the proposed method can calibrate the required accuracy to within 1.5mm and effectively shorten the set-up time, which took 5 hours per robot unit, to within 40 minutes. By using the developed system, it is possible to shorten the OLP working time of the car body assembly line, shorten the precision teaching time of the robot, improve the quality of the produced product and minimize the defect rate.

• Journal of the Korean Society of Radiology
• /
• v.13 no.2
• /
• pp.185-192
• /
• 2019

DXA is the most commonly used BMD examination equipment with the best performance on reflecting the biological alteration with tiny change of bone density. In spite of the importance of the quality control to maintain the accuracy and precision of the examination, considerable number of hospitals are not conducting QC due to the difficulty and high cost of the phantom product. This study develops the cross revision phantom with 3D printer and the change of the degree of infilling filaments which can be readily secured, and provides the usefulness assessment of the developed phantom by comparing with existing products. The Hounsfield Units of ABS, TPU, PLA, 30% Cu-PLA, and 30% Al-PLA are assessed. The Hounsfield Units result at infilling rate 100% was $-149.74{\pm}2.36$, $-55.62{\pm}7.14$, $-7.68{\pm}3.82$, $87.53{\pm}1.07$, and $1795.20{\pm}16.15$. The L1, L2, L3 BMD of 3D printing phantom with linear regression model were $0.620{\pm}0.010g/cm^2$, $1.092{\pm}0.025g/cm^2$, $1.554{\pm}0.026g/cm^2$ which are statistically relevant to the existing phantom products. This result provides the base line data for various medical phantom produce and capability of proper quality control of DXA equipment.

• Journal of Food Hygiene and Safety
• /
• v.34 no.4
• /
• pp.374-379
• /
• 2019

This study developed predictive growth models of Salmonella enterica Serovar Typhimurium on lettuce washed with chlorine (100~300 ppm) and ultrasound (US, 37 kHz, 380 W) treatment and stored at different temperatures ($10{\sim}25^{\circ}C$) using a polynomial equation. The primary model of specific growth rate (SGR) and lag time (LT) showed a good fit ($R^2{\geq}0.92$) with a Gompertz equation. A secondary model was obtained using a quadratic polynomial equation. The appropriateness of the secondary SGR and LT model was verified by coefficient of determination ($R^2=0.98{\sim}0.99$ for internal validation, 0.97~0.98 for external validation), mean square error (MSE=-0.0071~0.0057 for internal validation, -0.0118~0.0176 for external validation), bias factor ($B_f=0.9918{\sim}1.0066$ for internal validation, 0.9865~1.0205 for external validation), and accuracy factor ($A_f=0.9935{\sim}1.0082$ for internal validation, 0.9799~1.0137 for external validation). The newly developed models for S. Typhimurium could be incorporated into a tertiary modeling program to predict the growth of S. Typhimurium as a function of combined chlorine and US during the storage. These new models may also be useful to predict potential S. Typhimurium growth on lettuce, which is important for food safety purposes during the overall supply chain of lettuce from farm to table. Finally, the models may offer reliable and useful information of growth kinetics for the quantification microbial risk assessment of S. Typhimurium on washed lettuce.

• The Korean Journal of Archival Studies
• /
• no.45
• /
• pp.155-188
• /
• 2015

This study aims to find whether there is a difference between the resulting responsiveness and substantial responsiveness, as noticing citizen satisfaction of the result of information disclosure is not equivalent to a high rate of information disclosure. Previous studies focused on the analysis of the resulting responsiveness such as disclosure decision and processing time. However, this study would identify how much opened information is equal to requested information on the side of substantial responsiveness. This study found that accuracy dropped as opening not requested information but different information and completeness dropped as omitting some part of information or opening unseizable information on the side of substantial responsiveness. There are differences between the resulting responsiveness and substantial responsiveness. Some of the opened information is not requested information despite the disclosure decision. It takes over ten days despite the immediately disclose decision. The main reason for the decline in substantial responsiveness is the passing of document retention period and the absence of data. Therefore, the obligation for the creation and preservation of records for public agencies will have to be followed with the agencies' will to opening information. Although this study analyzes the limited cases, it is significant to enunciate there are differences between the resulting responsiveness and substantial responsiveness.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.45 no.1
• /
• pp.87-93
• /
• 2019

This study attempted to eatablish a High Performance Liquid Chromatography (HPLC) analysis method for the determination of (-)-epicatechin gallate as a part of the quality control for the development of functional cosmetic materials from Penthorum chinense Pursh. HPLC was performed on a Unison $US-C_{18}$ column ($4.6{\times}250mm$, $5{\mu}m$) with a gradient elution of 0.05% (v/v) trifluoroacetic acid (TFA) and methyl alcohol at a flow rate of 1.0 mL/min at $30^{\circ}C$. The analyte was detected at 280 nm. The HPLC method was performed in accordance with the International Conference on Harmonization (ICH) guideline (version 4, 2005) of analytical procedures with respect to specificity, precision, accuracy, and linearity. The limits of detection and quantitation were 0.11 and 0.33 mg/mL, respectively. Calibration curves showed good linearity ($r^2$ > 0.9999), and the precision of analysis was satisfied (less than 0.6%). Recoveries of quantified compounds ranged from 99.51 to 101.92%. This result indicates that the established HPLC method is very useful for the determination of marker compound in P. chinense Pursh extracts.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.12
• /
• pp.653-659
• /
• 2018

The accurate prediction of construction cost is a key factor in a project's success. However, it is hard to predict the construction costs in the planning stages rapidly and precisely when drawings, specifications, construction cost calculation statements are incomplete, among other factors. Accurate construction-cost prediction in the planning stage of a project is also important for project feasibility studies and successful completion. Therefore, various techniques have been applied to accurately predict construction costs at an early stage when project information is limited. There are many factors that affect the construction cost prediction. This paper presents a construction-cost prediction method as multiple regression model with seven construction factors as independent variables. The method was used to predict the construction cost of a local housing union project, and the error rate was 4.87%. It is not possible to compare the cost of the project at the planning stage of the local housing union project, but it has high prediction accuracy compared to the unit price of an existing unit area. It is likely to be applied in construction-cost calculation work and to contribute to the establishment of the budget for the local housing union project.

• Journal of Food Hygiene and Safety
• /
• v.34 no.3
• /
• pp.263-268
• /
• 2019

Methanol is a toxic alcohol used in various products such as antifreeze, detergent, disinfectant and industrial solvent. In the human body, methanol is oxidized to formaldehyde and formic acid, which can lead to metabolic acidosis, optic nerve impairment, and death. In this study, the methanol levels in detergents (n=191) and rinse aids (n=13) were analyzed by gas chromatography-headspace-mass spectrometry (GC-HS-MS). Limit of detection was 1.09 mg/kg, accuracy and precision were 91.1-97.9% and <10%, and it was suitable for quantitative analysis. This analysis method was simple and fast with a higher recovery rate than the conventional MFDS (Ministry of Food and Drug Safety) method of diluting the sample in water and putting it in a headspace vial.

• Journal of the Korea Convergence Society
• /
• v.10 no.5
• /
• pp.1-8
• /
• 2019

In this paper, we propose an improved facial skin color extraction method to solve the problem that facial surface is lost due to shadow or illumination in skin color extraction process and skin color extraction is not possible. In the conventional HSV method, when facial surface is brightly illuminated by light, the skin color component is lost in the skin color extraction process, so that a loss area appears on the face surface. In order to solve these problems, we extract the skin color, determine the elements in the H channel value range of the skin color in the HSV color space among the lost skin elements, and combine the coordinates of the lost part with the coordinates of the original image, To minimize the number of In the face detection process, the face was detected using the LBP Cascade Classifier, which represents texture feature information in the extracted skin color image. Experimental results show that the proposed method improves the detection rate and accuracy by 5.8% and 9.6%, respectively, compared with conventional RGB and HSV skin color extraction and face detection using the LBP cascade classifier method.

• Asian-Australasian Journal of Animal Sciences
• /
• v.32 no.7
• /
• pp.1015-1026
• /
• 2019

Objective: The objective of this study was to evaluate three different degrees of white striping (WS) addressing their automatic assessment and customer acceptance. The WS classification was performed based on a computer vision system (CVS), exploring different machine learning (ML) algorithms and the most important image features. Moreover, it was verified by consumer acceptance and purchase intent. Methods: The samples for image analysis were classified by trained specialists, according to severity degrees regarding visual and firmness aspects. Samples were obtained with a digital camera, and 25 features were extracted from these images. ML algorithms were applied aiming to induce a model capable of classifying the samples into three severity degrees. In addition, two sensory analyses were performed: 75 samples properly grilled were used for the first sensory test, and 9 photos for the second. All tests were performed using a 10-cm hybrid hedonic scale (acceptance test) and a 5-point scale (purchase intention). Results: The information gain metric ranked 13 attributes. However, just one type of image feature was not enough to describe the phenomenon. The classification models support vector machine, fuzzy-W, and random forest showed the best results with similar general accuracy (86.4%). The worst performance was obtained by multilayer perceptron (70.9%) with the high error rate in normal (NORM) sample predictions. The sensory analysis of acceptance verified that WS myopathy negatively affects the texture of the broiler breast fillets when grilled and the appearance attribute of the raw samples, which influenced the purchase intention scores of raw samples. Conclusion: The proposed system has proved to be adequate (fast and accurate) for the classification of WS samples. The sensory analysis of acceptance showed that WS myopathy negatively affects the tenderness of the broiler breast fillets when grilled, while the appearance attribute of the raw samples eventually influenced purchase intentions.