• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.1
• /
• pp.56-63
• /
• 2016

In this study, we introduce a design of Fuzzy RBFNNs-based digit recognition system using the incremental-PCA in order to recognize the handwritten digits. The Principal Component Analysis (PCA) is a widely-adopted dimensional reduction algorithm, but it needs high computing overhead for feature extraction in case of using high dimensional images or a large amount of training data. To alleviate such problem, the incremental-PCA is proposed for the computationally efficient processing as well as the incremental learning of high dimensional data in the feature extraction stage. The architecture of Fuzzy Radial Basis Function Neural Networks (RBFNN) consists of three functional modules such as condition, conclusion, and inference part. In the condition part, the input space is partitioned with the use of fuzzy clustering realized by means of the Fuzzy C-Means (FCM) algorithm. Also, it is used instead of gaussian function to consider the characteristic of input data. In the conclusion part, connection weights are used as the extended diverse types in polynomial expression such as constant, linear, quadratic and modified quadratic. Experimental results conducted on the benchmarking MNIST handwritten digit database demonstrate the effectiveness and efficiency of the proposed digit recognition system when compared with other studies.

• Information Systems Review
• /
• v.22 no.1
• /
• pp.59-72
• /
• 2020

Artificial Intelligence is establishing itself as a familiar tool from an intractable concept. In this trend, financial sector is also looking to improve the problem of existing system which includes Fraud Detection System (FDS). It is being difficult to detect sophisticated cyber financial fraud using original rule-based FDS. This is because diversification of payment environment and increasing number of electronic financial transactions has been emerged. In order to overcome present FDS, this paper suggests 3 types of artificial intelligence models, Generative Adversarial Network (GAN), Deep Neural Network (DNN), and Convolutional Neural Network (CNN). GAN proves how data imbalance problem can be developed while DNN and CNN show how abnormal financial trading patterns can be precisely detected. In conclusion, among the experiments on this paper, WGAN has the highest improvement effects on data imbalance problem. DNN model reflects more effects on fraud classification comparatively.

• Journal of Bio-Environment Control
• /
• v.23 no.3
• /
• pp.229-234
• /
• 2014

We analyzed drainage water from coir substrate in which cucumber plants were grown in winter and elucidated changes in pH, EC, and major nutrients according to the growth stages to recommend nutrient solution management appropriate to each growth stage. From the analysis of drainage solution the growth stages of cucumber were desirable to be divided into two, planting to fruit setting and fruit setting to harvest in case of nutrient solution management. The time required was about 3 weeks from planting to the first fruit setting and thereafter 7~10 days more until the first harvest. Approximately every 3~4 days were needed until the upper flowers bloomed. The time required from fruit setting to harvest was not different much among flowers as cucumber plants grew. From the experimental results, EC of supplied solution was recommended to maintain a little high to $3.0dS{\cdot}m^{-1}$ until before fruit setting and lower a little to $2.0{\sim}2.3dS{\cdot}m^{-1}$ after that. Of course, the amount of solution supply should be increased as plants grew. In case of each nutrients, the recommendation of concentrations of nitrogen, phosphorus and calcium were 700, 60, and $110mg{\cdot}L^{-1}$ each until before fruit setting, and then 660, 50, and $100mg{\cdot}L^{-1}$ each after fruit setting. The concentrations of potassium and magnesium are recommended to start from 400 and $80mg{\cdot}L^{-1}$ until fruit setting and lower a little after that.

• Nuclear Medicine and Molecular Imaging
• /
• v.41 no.4
• /
• pp.299-308
• /
• 2007

Purpose: The aim of this study was to develop a bioinformatics software and to test it in serum samples of papillary thyroid cancer using mass spectrometry (SELDI-TOF-MS). Materials and Methods: Development of 'Protein analysis' software performing decision tree analysis was done by customizing C4.5. Sixty-one serum samples from 27 papillary thyroid cancer, 17 autoimmune thyroiditis, 17 controls were applied to 2 types of protein chips, CM10 (weak cation exchange) and IMAC3 (metal binding - Cu). Mass spectrometry was performed to reveal the protein expression profiles. Decision trees were generated using 'Protein analysis' software, and automatically detected biomarker candidates. Validation analysis was performed for CM10 chip by random sampling. Results: Decision tree software, which can perform training and validation from profiling data, was developed. For CM10 and IMAC3 chips, 23 of 113 and 8 of 41 protein peaks were significantly different among 3 groups (p<0.05), respectively. Decision tree correctly classified 3 groups with an error rate of 3.3% for CM10 and 2.0% for IMAC3, and 4 and 7 biomarker candidates were detected respectively. In 2 group comparisons, all cancer samples were correctly discriminated from non-cancer samples (error rate = 0%) for CM10 by single node and for IMAC3 by multiple nodes. Validation results from 5 test sets revealed SELDI-TOF-MS and decision tree correctly differentiated cancers from non-cancers (54/55, 98%), while predictability was moderate in 3 group classification (36/55, 65%). Conclusion: Our in-house software was able to successfully build decision trees and detect biomarker candidates, therefore it could be useful for biomarker discovery and clinical follow up of papillary thyroid cancer.

• Journal of Nutrition and Health
• /
• v.50 no.1
• /
• pp.10-24
• /
• 2017

Purpose: GST (glutathione S-transferase) M1 and T1 gene polymorphisms are known to affect antioxidant levels. This study was carried out to evaluate genetic susceptibility by measuring the effect of DNA damage reduction in the Korean diet by vegetable food according to GST gene polymorphisms using the ex vivo method with human lymphocytes. Methods: Vegetable foods in the Korean diet based the results of the KNHANES V-2 (2011) were classified into 10 food groups. A total of 84 foods, which constituted more than 1% of the total intake in each food group, were finally designated as a vegetable food in the Korean diet. The Korean diet applied in this study is the standard one-week meals for Koreans (2,000 Kcal/day) suggested by the 2010 Dietary Reference Intakes for Koreans. Ex vivo DNA damage in human lymphocytes was assessed using comet assay. Results: In the Korean food group, the DNA damage protective effect of GSTM1 and GSTT1 was found to be greater in mutant type and wild-type, respectively. and the DNA damage protective effect according to the combined genotype of GSTM1 and GSTT1 was different depending on the food group. On the other hand, in Korean Diet, the DNA damage protective effect appeared to be larger in GSTM1 wild-type than in mutant type and was found to not be affected by GSTT1 genotype. Conclusion: These results can be used as basic data to demonstrate the superiority of the antioxidant function of Korean dietary patterns and food groups. Furthermore, it may be a starting point to begin research on customized antioxidant nutrition according to individual genes.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.50 no.3
• /
• pp.318-333
• /
• 2023

The purpose of this study is to analyze changes in dental trauma in children under the age of 12 during the period of Coronavirus Disease 2019 (COVID-19). March 2020, when COVID-19 was officially declared a pandemic by the World Health Organization, was set as the starting point for COVID-19. From March 2018 to February 2020, subjects in the pre-COVID-19 period were classified as the Pre-COVID-19 group, and from March 2020 to March 2022, subjects in the post-COVID-19 period were classified as the COVID-19 group. Information related to trauma was collected through electronic medical records. The number of trauma patients before and after the outbreak of COVID-19 decreased significantly. During the COVID-19 period, there was no significant difference in the male-female ratio or the distribution order of age groups. In the COVID-19 group of permanent teeth, the ratio of trauma caused by personal mobility was higher than trauma caused by sports. In the COVID-19 group of permanent teeth, the ratio of crown fracture with pulp involvement was significantly higher than the ratio of crown fracture without pulp involvement. Changes in trauma patterns caused by COVID-19 were observed more clearly in school-aged children than in preschool children. In a pandemic situation such as COVID-19, it is expected to be used as a good educational basis for knowing that frequent diagnoses can change due to changes in the environment.

• Journal of The Korean Association For Science Education
• /
• v.31 no.1
• /
• pp.78-98
• /
• 2011

The purpose of this study was to analyze elementary school students' interpretation of data characteristics by cognitive style. Participants were elementary students in sixth grade who can use integrated inquiry process skills. The students were divided into two groups, analytic cognitive style and wholistic cognitive style according to their response to Cognitive Style Analysis. They performed scientific interpretation of data activity. To collect data for this study, participants recorded the result on scientific interpretation of data activity paper and researcher recorded the situation on videotape and interviewed with participants after the end of interpretation of data to get additional data. And the findings of this study were as follows: First, the study analyzed interpretation of data characteristics by the operator regarding different situations of interpreting data according to cognitive style. For example, in the intermediate state, analytic-cognitive style students showed high achievement in identifying variables, and wholistic-cognitive style students were active in using prior knowledge to interpret data. Second, the result of analysis on the direction of interpreting data and preference for data types in interpreting data activities according to cognitive style are as follows: Wholistic-cognitive style students showed relatively high perception of information through the top-down approach. On the other hand, analytic-cognitive style students usually used the bottom-up approach gradually expanding detailed information to the scientific question-related answer and showed a preference data of the table type. Through the result, this study aimed to help establish a data interpretation strategy for learners to solve problems based on understanding of interpretation of data characteristics according to learners' cognitive style, and purposed the instruction design suggesting the data requiring various data interpretation strategies to develop learners' data interpretation ability.

• Economic and Environmental Geology
• /
• v.55 no.3
• /
• pp.241-259
• /
• 2022

In general, the REE were produced by mining conventional deposits, such as the carbonatite or the clay-hosted REE deposits. However, because of the recent demand increase for REE in modern industries, unconventional REE deposits emerged as a necessary research topic. Among the unconventional REE recovery methods, the REE-bearing coal deposits are recently receiving attentions. R-types generally have detrital originations from the bauxite deposits, and show LREE enriched REE patterns. Tuffaceous-types are formed by syngenetic volcanic activities and following input of volcanic ash into the basin. This type shows specific occurrence of the detrital volcanic ash-driven minerals and the authigenic phosphorous minerals focused at narrow horizon between coal seams and tonstein layers. REE patterns of tuffaceous-types show flat shape in general. Hydrothermal-types can be formed by epigenetic inflow of REE originated from granitic intrusions. Occurrence of the authigenic halogen-bearing phosphorous minerals and the water-bearing minerals are the specific characteristics of this type. They generally show HREE enriched REE patterns. Each type of REE-bearing coal deposits may occur by independent genesis, but most of REE-bearing coal deposits with high REE concentrations have multiple genesis. For the case of the US, the rare earth oxides (REO) with high purity has been produced from REE-bearing coals and their byproducts in pilot plants from 2018. Their goal is to supply about 7% of national REE demand. For the coal deposits in Korea, lignite layers found in Gyungju-Yeongil coal fields shows coexistence of tuff layers and coal seams. They are also based in Tertiary basins, and low affection from compaction and coalification might resulted into high-REE tuffaceous-type coal deposits. Thus, detailed geologic researches and explorations for domestic coal deposits are required.

• Radiation Oncology Journal
• /
• v.23 no.2
• /
• pp.61-70
• /
• 2005

Purpose: Uterine cervix cancer is one of the most prevalent women cancer in Korea. We analysed published papers in Korea with comparing Patterns of Care Study (PCS) articles of United States and Japan for the purpose of developing and processing Korean PCS. Materials and Methods: We searched PCS related foreign-produced papers in the PCS homepage (212 articles and abstracts) and from the Pub Med to find Structure and Process of the PCS. To compare their study with Korean papers, we used the internet site 'Korean Pub Med' to search 99 articles regarding uterine cervix cancer and radiation therapy. We analysed Korean paper by comparing them with selected PCS papers regarding Structure, Process and Outcome and compared their items between the period of before 1980's and 1990's. Results: Evaluable papers were 28 from United States, 10 from the Japan and 73 from the Korea which treated cervix PCS items. PCS papers for United States and Japan commonly stratified into $3\~4$ categories on the bases of the scales characteristics of the facilities, numbers of the patients, doctors, Researchers restricted eligible patients strictly. For the process of the study, they analysed factors regarding pretreatment staging in chronological order, treatment related factors, factors in addition to FIGO staging and treatment machine. Papers in United States dealt with racial characteristics, socioeconomic characteristics of the patients, tumor size (6), and bilaterality of parametrial or pelvic side wail invasion (5), whereas papers from Japan treated of the tumor markers. The common trend in the process of staging work-up was decreased use of lymphangiogram, barium enema and increased use of CT and MRI over the times. The recent subject from the Korean papers dealt with concurrent chemoradiotherapy (9 papers), treatment duration (4), tumor markers (B) and unconventional fractionation. Conclusion: By comparing papers among 3 nations, we collected items for Korean uterine cervix cancer PCS. By consensus meeting and close communication, survey items for cervix cancer PCS were developed to measure structure, process and outcome of the radiation treatment of the cervix cancer. Subsequent future research will focus on the use of brachytherapy and its impact on outcome including complications. These finding and future PCS studies will direct the development of educational programs aimed at correcting identified deficits in care.