• 한국HCI학회:학술대회논문집
• /
• 2008.02b
• /
• pp.370-375
• /
• 2008

When customer makes a phone call to get the necessary information, ARS (Automatic Response System) will provide them without the limitation of time and place. Therefore, the application field of ARS (Automatic Response System) has been expanded and also occupancy rates arepretty high. However, compared with the importance of ARS, there is little study regarding the closest information structure of the actual users. Therefore, this research is about the ARS information structure of Gimhea BIS (Bus Information System), and the point of this research is to analyze the problem of the ARS information structure so we can suggest an idea to improve it. As the methods of study, drawing the workflow of ARS, conducting user observation, comparing with cases another region and analysis them. As a result, by and large there was error in the structure of information so, system reformation was required. Now that, we suggested two ideas for the improvement and we conducted usability test under that problem. Usability test was conducted by Lab test with interview. As a result, we could understand merits and demerits of two ideas for improvement, and we completed final improvement to highlight only merits of two ideas for improvement. This study will be utilized as the base material to improve ARS of Gimhea BIS.

• Journal of KIISE:Software and Applications
• /
• v.33 no.8
• /
• pp.668-680
• /
• 2006

Web services on the Internet are not always reliable in terms of service availability and performance. Dynamic service coordination capability of a system or an application invoking Web services is essential to cope with such unreliable situations. In dynamic service coordination, if a Web service does not respond within a specific time constraint, it is replaced with another Web service at run time for reliable invocation of Web services. In this paper, we develop an exception based dynamic service coordination framework for Web services. In the framework, all necessary information for dynamic service coordination is explicitly specified and summarized as a set of attributes. Then classes and workflows, supporting dynamic service coordination and invoking Web services, are automatically created based on these attributes. Developers of Web services client programs can make the invocations of Web services reliable by calling the methods of the classes. Some performance loss has been observed in the indirect invocation of a Web service. However, when we consider the flexibility and reliability gained from the method, the performance loss would be acceptable in many cases.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.5
• /
• pp.1273-1284
• /
• 2016

In this paper, we analyzed the patents on machine learning using keyword network analysis and clique analysis. To construct a keyword network, important keywords were extracted based on the TF-IDF weight and their association, and network structure analysis and clique analysis was performed. Density and clustering coefficient of the patent keyword network are low, which shows that patent keywords on machine learning are weakly connected with each other. It is because the important patents on machine learning are mainly registered in the application system of machine learning rather thant machine learning techniques. Also, our results of clique analysis showed that the keywords found by cliques in 2005 patents are the subjects such as newsmaker verification, product forecasting, virus detection, biomarkers, and workflow management, while those in 2015 patents contain the subjects such as digital imaging, payment card, calling system, mammogram system, price prediction, etc. The clique analysis can be used not only for identifying specialized subjects, but also for search keywords in patent search systems.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.5
• /
• pp.305-315
• /
• 2016

Due to the change from small volume production to small quantity batch production systems, individual companies have been attempting to produce a wide range of operating strategies, maximize their productivity, and minimize their WIP level by operating with the proper cycle time to defend their market share. In particular, using a complex workflow and process sequence in the manufacturing line has some drawbacks when it comes to designing the production strategy by applying analytical models, such as mathematical models and queueing theory. For this purpose, this paper uses three heuristic algorithms to solve the job release problem at the bottleneck workstation, product mix problem in multi-purpose machine(s), and batch size and sequence in batch machine(s). To verify the effectiveness of the proposed methods, a simulation analysis was performed. The experimental results demonstrated that the combined application of the proposed methods showed positive effects on the reduction of the cycle time and WIP level, and improvement of the throughput.

• Journal of the Korean Academy of Esthetic Dentistry
• /
• v.27 no.2
• /
• pp.82-96
• /
• 2018

3D printing is a process of producing 3d object from a digital file in STL format by joining, bonding, sintering or polymerizing small volume elements by layer. The various type of 3d printing is classified according to the additive manufacturing strategies. Among the types of 3D printer, SLA(StereoLithography Apparatus) and DLP(Digital Light Processing) 3D printer which use polymerization by light source are widely used in dental office. In the previous study, a full-arch scale 3d printed model is less precise than a conventional stone model. However, in scale of quadrant arch, a 3d printed model is significantly precise than a five-axis milled model. Using $3^{rd}$ Party dental CAD program, full denture, provisional crowns and diagnostic wax-up model are fabricated by 3d printer in dental office. In Orthodontics, based on virtual setup model, indirect bracket bonding tray can be generated by 3d printer. And thermoforming clear aligner can be fabricated on the 3d printed model. 3D printed individual drilling guide enable the clinician to place the dental implant on the proper position. The development of layer additive technology enhance the quality of 3d printing object and shorten the operating time of 3D printing. In the near future, traditional dental laboratory process such as casting, denture curing will be replaced by digital 3D printing.

• The Journal of Advanced Prosthodontics
• /
• v.13 no.6
• /
• pp.361-372
• /
• 2021

PURPOSE. To improve the clinical effects of complete denture use and simplify its clinical application, a digital complete denture restoration workflow (Functional Suitable Digital Complete Denture System, FSD) was proposed and preliminary clinical evaluation was done. MATERIALS AND METHODS. Forty edentulous patients were enrolled, of which half were treated by a prosthodontic chief physician, and the others were treated by a postgraduate student. Based on the primary impression and jaw relation obtained at the first visit, diagnostic denture was designed and printed to create a definitive impression, jaw relation, and esthetic confirmation at the second visit. A redesigned complete denture was printed as a mold to fabricate final denture that was delivered at the third visit. To evaluate accuracy of impression made by diagnostic denture, the final denture was used as a tray to make impression, and 3D comparison was used to analyze their difference. To evaluate the clinical effect of FSD, visual analogue scores (VAS) were determined by both dentists and patients. RESULTS. Two visits were reduced before denture delivery. The RMS values of 3D comparison between the impression made via diagnostic dentures and the final dentures were 0.165 ± 0.033 mm in the upper jaw and 0.139 ± 0.031 mm in the lower jaw. VAS ratings were between 8.5 and 9.6 in the chief physician group, while 7.7 and 9.5 in the student group; there was no statistical difference between the two groups. CONCLUSION. FSD can simplify the complete denture restoration process and reduce the number of visits. The accuracy of impressions made by diagnostic dentures was acceptable in clinic. The VASs of both dentists and patients were satisfied.

• Journal of Internet Computing and Services
• /
• v.24 no.4
• /
• pp.127-135
• /
• 2023

This research paper explores the application of FinBERT, a variational BERT-based model pre-trained on financial domain, for sentiment analysis in the financial domain while focusing on the process of identifying suitable training data and hyperparameters. Our goal is to offer a comprehensive guide on effectively utilizing the FinBERT model for accurate sentiment analysis by employing various datasets and fine-tuning hyperparameters. We outline the architecture and workflow of the proposed approach for fine-tuning the FinBERT model in this study, emphasizing the performance of various datasets and hyperparameters for sentiment analysis tasks. Additionally, we verify the reliability of GPT-3 as a suitable annotator by using it for sentiment labeling tasks. Our results show that the fine-tuned FinBERT model excels across a range of datasets and that the optimal combination is a learning rate of 5e-5 and a batch size of 64, which perform consistently well across all datasets. Furthermore, based on the significant performance improvement of the FinBERT model with our Twitter data in general domain compared to our news data in general domain, we also express uncertainty about the model being further pre-trained only on financial news data. We simplify the complex process of determining the optimal approach to the FinBERT model and provide guidelines for selecting additional training datasets and hyperparameters within the fine-tuning process of financial sentiment analysis models.

• Journal of Genetic Medicine
• /
• v.7 no.1
• /
• pp.59-66
• /
• 2010

Purpose: Quantitative fluorescent polymerase chain reaction (QF-PCR) allows for the rapid prenatal diagnosis of common aneuploidies. The main advantages of this assay are its low cost, speed, and automation, allowing for large-scale application. However, despite these advantages, it is not a routine method for prenatal aneuploidy screening in Korea. Our objective in the present study was to validate the performance of QF-PCR using short tandem repeat (STR) markers in a Korean population as a means for rapid prenatal diagnosis. Material and Methods: A QF-PCR assay using an Elucigene kit (Gen-Probe, Abingdon, UK), containing 20 STR markers located on chromosomes 13, 18, 21, X and Y, was performed on 847 amniotic fluid (AF) samples for prenatal aneuploidy screening referred for prenatal aneuploidy screening from 2007 to 2009. The results were then compared to those obtained using conventional cytogenetic analysis. To evaluate the informativity of STR markers, the heterozygosity index of each marker was determined in all the samples. Results: Three autosomes (13, 18, and 21) and X and Y chromosome aneuploidies were detected in 19 cases (2.2%, 19/847) after QF-PCR analysis of the 847 AF samples. Their results are identical to those of conventional cytogenetic analysis, with 100% positive predictive value. However, after cytogenetic analysis, 7 cases (0.8%, 7/847) were found to have 5 balanced and 2 unbalanced chromosomal abnormalities that were not detected by QF-PCR. The STR markers had a slightly low heterozygosity index (average: 0.76) compared to those reported in Caucasians (average: 0.80). Submicroscopic duplication of D13S634 marker, which might be a unique finding in Koreans, was detected in 1.4% (12/847) of the samples in the present study. Conclusion: A QF-PCR assay for prenatal aneuploidy screening was validated in our institution and proved to be efficient and reliable. However, we suggest that each laboratory must perform an independent validation test for each STR marker in order to develop interpretation guidelines of the results and must integrate QF-PCR into the routine cytogenetic laboratory workflow.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.2
• /
• pp.81-90
• /
• 2013

During the past decade, many changes and attempts have been tried and are continued developing new technologies in the computing area. The brick wall in computing area, especially power wall, changes computing paradigm from computing hardwares including processor and system architecture to programming environment and application usage. The high performance computing (HPC) area, especially, has been experienced catastrophic changes, and it is now considered as a key to the national competitiveness. In the late 2000's, many leading countries rushed to develop Exascale supercomputing systems, and as a results tens of PetaFLOPS system are prevalent now. In Korea, ICT is well developed and Korea is considered as a one of leading countries in the world, but not for supercomputing area. In this paper, we describe architecture design of MAHA supercomputing system which is aimed to develop 300 TeraFLOPS system for bio-informatics applications like human genome analysis and protein-protein docking. MAHA supercomputing system is consists of four major parts - computing hardware, file system, system software and bio-applications. MAHA supercomputing system is designed to utilize heterogeneous computing accelerators (co-processors like GPGPUs and MICs) to get more performance/$, performance/area, and performance/power. To provide high speed data movement and large capacity, MAHA file system is designed to have asymmetric cluster architecture, and consists of metadata server, data server, and client file system on top of SSD and MAID storage servers. MAHA system softwares are designed to provide user-friendliness and easy-to-use based on integrated system management component - like Bio Workflow management, Integrated Cluster management and Heterogeneous Resource management. MAHA supercomputing system was first installed in Dec., 2011. The theoretical performance of MAHA system was 50 TeraFLOPS and measured performance of 30.3 TeraFLOPS with 32 computing nodes. MAHA system will be upgraded to have 100 TeraFLOPS performance at Jan., 2013.