• Journal of Applied Biological Chemistry
• /
• v.66
• /
• pp.394-403
• /
• 2023

In the pursuit of sustainable and environmentally-friendly agricultural practices, we conducted an extensive study on the rhizosphere bacteria inhabiting soils that have been devoid of chemical fertilizers for an extended period exceeding 40 years. Through this investigation, we isolated a total of 80 species of plant growth-promoting rhizosphere bacteria and assessed their potential to enhance plant growth. Among these isolates, Burkholderia cepacia CD2 displayed remarkable plant growth-promoting activity, making it an optimal candidate for further analysis. Burkholderia cepacia CD2 exhibited a range of beneficial characteristics conducive to plant growth, including phosphate solubilization, siderophore production, denitrification, nitrate utilization, and urease activity. These attributes are well-known to positively influence the growth and development of plants. To validate the taxonomic classification of the strain, 16S rRNA gene sequencing confirmed its placement within the Burkholderia genus, providing further insights into its phylogenetic relationship. To delve deeper into the potential mechanisms underlying its plant growth-promoting properties, we sought to confirm the presence of specific genes associated with plant growth promotion in CD2. To achieve this, whole genome sequencing (WGS) was performed by Plasmidsaurus Inc. (USA) utilizing Oxford Nanopore technology (Abingdon, UK). The WGS analysis of the genome of CD2 revealed the existence of a subsystem function, which is thought to be a pivotal factor contributing to improved plant growth. Based on these findings, it can be concluded that Burkholderia cepacia CD2 has the potential to serve as a microbial fertilizer, offering a sustainable alternative to chemical fertilizers.

• Journal of Korean Library and Information Science Society
• /
• v.55 no.2
• /
• pp.29-53
• /
• 2024

Public library programs serve a cultural function and are a tool for community interaction. Recently, there has been an increase in the number of websites that integrate the programs of various institutions, but there is no set framework to describe the programs. Therefore, in order to prepare a framework for program information, we conducted a content analysis of Jeongdok Library programs as an example. Using MAXQDA, a content analysis tool, category codes for type, topic, special classification, and target audience were derived. Based on this, we analyzed the characteristics of the Jeongdok library programs as follows. In terms of type, there are many programs such as classes and lectures, but fewer programs related to tours, performances and screenings, and operational experiences. In terms of topic matter, programs related to reading and the arts were dominant, while programs related to book curation, awards, and the environment were less common. In terms of target audience, the most common programs were for adults, with fewer programs for high school students, middle school students, and library staff. The framework of Jeongdok library program can be used not only to classify the programs currently operated by other public libraries, but also to develop a service platform for public library programs in Korea.

• Journal of the Korean Society of Radiology
• /
• v.85 no.4
• /
• pp.714-726
• /
• 2024

Researchers have developed various algorithms utilizing artificial intelligence (AI) to automatically and objectively diagnose patterns and extent of pulmonary emphysema or interstitial lung diseases on chest CT scans. Studies show that AI-based quantification of emphysema on chest CT scans reveals a connection between an increase in the relative percentage of emphysema and a decline in lung function. Notably, quantifying centrilobular emphysema has proven helpful in predicting clinical symptoms or mortality rates of chronic obstructive pulmonary disease. In the context of interstitial lung diseases, AI can classify the usual interstitial pneumonia pattern on CT scans into categories like normal, ground-glass opacity, reticular opacity, honeycombing, emphysema, and consolidation. This classification accuracy is comparable to chest radiologists (70%-80%). However, the results generated by AI are influenced by factors such as scan parameters, reconstruction algorithms, radiation doses, and the training data used to develop the AI. These limitations currently restrict the widespread adoption of AI for quantifying pulmonary emphysema and interstitial lung diseases in daily clinical practice. This paper will showcase the authors' experience using AI for diagnosing and quantifying emphysema and interstitial lung diseases through case studies. We will primarily focus on the advantages and limitations of AI for these two diseases.

• Journal of Korea Water Resources Association
• /
• v.57 no.9
• /
• pp.615-626
• /
• 2024

To accurately and efficiently monitor soil moisture (SM) across South Korea, this study developed a SM estimation model that integrates the cloud computing platform Google Earth Engine (GEE) and Automated Machine Learning (AutoML). Various spatial information was utilized based on Terra MODIS (Moderate Resolution Imaging Spectroradiometer) and the global precipitation observation satellite GPM (Global Precipitation Measurement) to test optimal input data combinations. The results indicated that GPM-based accumulated dry-days, 5-day antecedent average precipitation, NDVI (Normalized Difference Vegetation Index), the sum of LST (Land Surface Temperature) acquired during nighttime and daytime, soil properties (sand and clay content, bulk density), terrain data (elevation and slope), and seasonal classification had high feature importance. After setting the objective function (Determination of coefficient, R² ; Root Mean Square Error, RMSE; Mean Absolute Percent Error, MAPE) using AutoML for the combination of the aforementioned data, a comparative evaluation of machine learning techniques was conducted. The results revealed that tree-based models exhibited high performance, with Random Forest demonstrating the best performance (R² : 0.72, RMSE: 2.70 vol%, MAPE: 0.14).

• Journal of Bio-Environment Control
• /
• v.33 no.1
• /
• pp.1-11
• /
• 2024

In this study, five artificial intelligence (AI) models: Inception v3, SqueezeNet (local), VGG-16, Painters, and DeepLoc were used to classify tea leaf diseases. Eight image categories were used: healthy, algal leaf spot, anthracnose, bird's eye spot, brown blight, gray blight, red leaf spot, and white spot. Software used in this study was Orange 3 which functions as a Python library for visual programming, that operates through an interface that generates workflows to visually manipulate and analyze the data. The precision of each AI model was recorded to select the ideal AI model. All models were trained using the Adam solver, rectified linear unit activation function, 100 neurons in the hidden layers, 200 maximum number of iterations in the neural network, and 0.0001 regularizations. To extend the functionality of Orange 3, new add-ons can be installed and, this study image analytics add-on was newly added which is required for image analysis. For the training model, the import image, image embedding, neural network, test and score, and confusion matrix widgets were used, whereas the import images, image embedding, predictions, and image viewer widgets were used for the prediction. Precisions of the neural networks of the five AI models (Inception v3, SqueezeNet (local), VGG-16, Painters, and DeepLoc) were 0.807, 0.901, 0.780, 0.800, and 0.771, respectively. Finally, the SqueezeNet (local) model was selected as the optimal AI model for the detection of tea diseases using tea leaf images owing to its high precision and good performance throughout the confusion matrix.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.5
• /
• pp.163-171
• /
• 2024

This study applies neural network analysis techniques to examine the factors influencing the purchasing decisions of fashion eyewear among women in their 30s and 40s, comparing these findings with traditional parametric analysis methods. In the fashion area, machine learning techniques are utilized for personalized fashion recommendation systems. However, research on such applications in Korea remains insufficient. By reanalyzing a study conducted in 2017 using traditional quantitative methods with these new techniques, this study aims to confirm the utility of neural network methods. Notably, the study finds that the classification accuracy of preferred sunglasses design is highest, at 86.2%, when the L-BFGS-B neural network is activated using the hyperbolic tangent function. The most critical factors influencing purchasing decisions were consumers' occupations and their pursuit of new styles. It is interpreted that Korean sunglasses consumers prefer "safe changes." These findings are consistent for selecting both the frames and lenses of sunglasses. Traditional quantitative analysis suggests that the type of sunglasses preferred varies according to the group to which a consumer belongs. In contrast, neural network analysis predicts the preferred sunglasses for each individual, thereby facilitating the development of personalized sunglasses recommendation systems.

• Journal of agricultural medicine and community health
• /
• v.20 no.2
• /
• pp.157-168
• /
• 1995

This study was carried out, through analyzing the annual reports(year of 1973-1993) on health status of Su Dong-Myun, and specific survey data of 332 households(Su Dong-Myun 209, Byul Nae-Myun 123), located in Nam Yang Ju-Si, Kyung Gi-Do, from July 20 to July 31, 1995, to find out more effective means for primary medical care in a rural area. The results were as fellows : 1. Number of population in Su Dong-Myun was 5,419 in 1973, 4,591(the lowest) in 1987 and 5,707 in 1995. In the composition rate of population, "0-14" of age group showed markedly decreasing tendency from 43.1% in 1975, to 19.1% in 1995, however "65 and over" markedly in creasing tendency form 5.3% in 1975 to 9.8% in 1995. 2. Annual utilization rate per 1,000 inhabitants in Su Dong-Myun showed markedly increasing tendency from 1973 to 1977 such as 343 in 1973, 540 in 1975, 900 in 1977. However, since 1979, the rate showed rapidly decreasing tendency, such as 846 in 1979, 519 in 1985, 190 in 1991 and 1993. 3. The morbid household rate per year was 53.6% of respondents and the rate per 15 days was 48.2%. In disease classification rate of morbid household per year, Arthralgia & Neuralgia was the highest rate(33.9%) and gastro-intestinal disorder(19.3%), Cough(11,9%), Hypertension(7.8%), Accident(3.2%) in next order. 4. In the utilizing facilities for Primary Medical Care, Medical facilities was showed the highest rate(58.1% of respondents) and Pharmacy and Drug Shp(33.1%), Tradition Method(4.0%) in next order. In the Medical facilities, General private clinic was showed the highest rate(34.3%) and specific private Clinic(22.3%), Hospital(19.0%), Health (Sub)center(16.3%), Nurse practitioner (3.3%), Oriental hospital and clinic(2.7%) in next order. 5. Experience rate, utilizing health subcenter was 51.8% of the respondents, and it was 55.0% in Su Dong-Myun and 46.3% in Byul Nae-Myun. In utilization times of health subcenter, times-rate showed next orders such as 1-2 times/6months(31.6%), 1-2 times/year (22.1%), 1-2 times/months(19.2%), 1-2 times/3months(15.6%). 6. In objectives, visiting Health Subcenter, Medical Care was the highest rate(59.8% of the respondents) and health control(23.3%) was in next order. In Medical Care, Primary Care by general physician was higher rate(51.1%) almost all. In the Health control, Immunization too was high rate(18.0%) in health control activities. 7. The reasons rate, utilizing health subcenter showed next order, such as distance to Medical facilities(33.0% of the respondents), Medical Cost(28.1%), Simple process of consultation (10.8%), Effectiveness of cure(7.6%), Function of primary medical care(7.0%) and Attitude of physician(6.5%). 8. In the affecting factors to utilization of primary medical facilities, medical needs was showed the highest rate(29.5% of the respondents) and medical cost(15.4%), distance to medical facilities(14.2%), traffic vehicle(14.2%) and farm work(6.9%) in next order. 9. In the priority between 'daily farm work,' and 'primary medical care', only 46.4% of respondents answered that primary health care is more important than the daily farm work The 22.6% of respondents answered 'daily farm work', and the 12.3% answered 'the equal of the both'. 10. In the criterion of medical facilities choice, medical knowledge and technical quality was showed the highest rate(56.3%), distance or time to medical facilities(10.9%), sincerity and kindness of physician(9.4%), medical cost(8.7%) and traffic vehicle(6.5%) in next order 11. In the advise for improvement of health subcenter function, the 36.1% of respondents answered that 'enforcement of medical personnel and equipment' was required, and then 'improved medical technology'(25.5%), 'good attitude of physician'(14.9%), 'improved medical system'(13.3%), 'enforced drug'(6.7%) in next order. 12. The study on affecting factors to utilization of primary medical facilities was very difficult subject to systematize the analyzed results, due to a prejudice of protocol planner, surveyer and respondent, and variety and overlapping of subject matter.

• Journal of Intelligence and Information Systems
• /
• v.20 no.4
• /
• pp.121-139
• /
• 2014

Predicting corporate failure has been an important topic in accounting and finance. The costs associated with bankruptcy are high, so the accuracy of bankruptcy prediction is greatly important for financial institutions. Lots of researchers have dealt with the topic associated with bankruptcy prediction in the past three decades. The current research attempts to use ensemble models for improving the performance of bankruptcy prediction. Ensemble classification is to combine individually trained classifiers in order to gain more accurate prediction than individual models. Ensemble techniques are shown to be very useful for improving the generalization ability of the classifier. Bagging is the most commonly used methods for constructing ensemble classifiers. In bagging, the different training data subsets are randomly drawn with replacement from the original training dataset. Base classifiers are trained on the different bootstrap samples. Instance selection is to select critical instances while deleting and removing irrelevant and harmful instances from the original set. Instance selection and bagging are quite well known in data mining. However, few studies have dealt with the integration of instance selection and bagging. This study proposes an improved bagging ensemble based on instance selection using genetic algorithms (GA) for improving the performance of SVM. GA is an efficient optimization procedure based on the theory of natural selection and evolution. GA uses the idea of survival of the fittest by progressively accepting better solutions to the problems. GA searches by maintaining a population of solutions from which better solutions are created rather than making incremental changes to a single solution to the problem. The initial solution population is generated randomly and evolves into the next generation by genetic operators such as selection, crossover and mutation. The solutions coded by strings are evaluated by the fitness function. The proposed model consists of two phases: GA based Instance Selection and Instance based Bagging. In the first phase, GA is used to select optimal instance subset that is used as input data of bagging model. In this study, the chromosome is encoded as a form of binary string for the instance subset. In this phase, the population size was set to 100 while maximum number of generations was set to 150. We set the crossover rate and mutation rate to 0.7 and 0.1 respectively. We used the prediction accuracy of model as the fitness function of GA. SVM model is trained on training data set using the selected instance subset. The prediction accuracy of SVM model over test data set is used as fitness value in order to avoid overfitting. In the second phase, we used the optimal instance subset selected in the first phase as input data of bagging model. We used SVM model as base classifier for bagging ensemble. The majority voting scheme was used as a combining method in this study. This study applies the proposed model to the bankruptcy prediction problem using a real data set from Korean companies. The research data used in this study contains 1832 externally non-audited firms which filed for bankruptcy (916 cases) and non-bankruptcy (916 cases). Financial ratios categorized as stability, profitability, growth, activity and cash flow were investigated through literature review and basic statistical methods and we selected 8 financial ratios as the final input variables. We separated the whole data into three subsets as training, test and validation data set. In this study, we compared the proposed model with several comparative models including the simple individual SVM model, the simple bagging model and the instance selection based SVM model. The McNemar tests were used to examine whether the proposed model significantly outperforms the other models. The experimental results show that the proposed model outperforms the other models.

• Journal of the Korean earth science society
• /
• v.26 no.7
• /
• pp.732-743
• /
• 2005

The characteristics of precipitation over South Korea from 1973 to 2002 were investigated. The synoptic patterns inducing precipitation are classified by 10 categories, according to the associated surface map analysis. The annual mean frequency of the total precipitation, its duration time and amount for 30 years are 179 times, 2.9 hours, and 7.1 mm, respectively. About $59\%$ of the total precipitation events were associated with a synoptic low. The dominant patterns are identified with respect to seasons: A synoptic mobile low pressure pattern is frequent in spring, fall, and winter, whereas low pressure embedded within the Changma and orography induced precipitation are dominant in summer and in winter. For the amount of precipitation, precipitation originated from tropical air associated with typhoon, tropical convergence, and Changma is more significant than that with other pressure patterns. The statistical elapse time reaching to 80 mm, which is the threshold amount of heavy rainfall watch at KMA, takes 12.9 hours after the onset of precipitation. The probability distribution function of the precipitation shows that the maximum probability for heavy rainfall is located at the south-coastal region of the Korean peninsula. It is also shown that the geographical distribution of the Korean peninsula plays an important role in occurrence of heavy rainfall. For example, heavy precipitation is frequently occurred at Youngdong area, when typhoon passes along the coastal region of the back borne mountains in the peninsula. The climatological classification of synoptic patterns associated with heavy rainfall over South Korea can be used to provide a guidance to operational forecast of heavy rainfall in KMA.

• Korean Journal of Biological Psychiatry
• /
• v.5 no.2
• /
• pp.253-262
• /
• 1998

Objectives : Risperidone, an atypical antipsychitics which blocks both dopaminergic and serotonergic receptors, have a good response to the negative symptoms as well as positive symptoms, and improve cognitive dysfunction of schizophrenic patients. Furthermore, it has few extrapyramidal side effects and tardive dyskinesia. Although it had been reported that the atypical antipsychotics have less effect on prolactin(PRL) than the classical antipsychotics, we could experience PRL-associated symptoms such as amenorrhea, galactorrhea and hyperprolactinemia in practice. Therefore, we tried to identify the sex differences of risperidone-induced hyperprolactinemia, to evaluate factors affecting PRL levels, and to know the association between cognitive disorders and PRL. Methods : The baseline levels of PRL and TSH prior to risperidone administration were measured by enzyme immunoassay method for 50 patients(25 ma-les and 25 females) admitted with schizophrenia, schizoaffective disorder or schizophreniform disorder according to the DSM-IV classification, and the measurements of PRL were repeated on the 2nd and the 4th wks of risperidone administration. Concomitantly, the severity of psychotic symptoms using CGI, BPRS and PANSS, and the cognitive dysfunction using PANSS-CF were assessed prior to, on the 2nd and the 4th wks of risperidone administration. The PRL and TSH levels of 54 healthy controls(29 males and 25 females) who had no medical, neurological and psychiatric illnesses were also evaluated. Furthermore, the correlation with the psychiatric diagnosis, education, age, sex, duration of illnesses, risperidone dosage, duration of risperdone administration, TSH concentration, cognitive function, severity of psychotic symptoms were also identified. Results : 1) The baseline PRL levels of female schizophrenics($74.3{\pm}49.6ng/ml$) were significantly(p<0.005) higher than those of males($36.3{\pm}24.6ng/ml$), which were significantly(p<0.0001 respectively) higher than those of controls(females $16.9{\pm}6.1ng/ml$, males $13.3{\pm}4.9ng/ml$). The PRL levels measured on the 2nd wks(females $133.7{\pm}47.8ng/ml$, males $56.9{\pm}23.6ng/ml$) and on the 4th wks(females $146.1{\pm}45.9ng/ml$, males $70.0{\pm}31.5ng/ml$) after risperidone administration were significantly(p<0.0001 respectively) higher in females. The mean dosages of risperidone on the 2nd wks were $3.8{\pm}1.7mg$(2-6mg) for the females and $4.0{\pm}1.6mg$(2-6mg) for the males, and on the 4th wks were $4.5{\pm}2.1mg$(2-8mg) for the females and $5.4{\pm}2.2mg$(2-8mg) for the males. 2) The rise of PRL levels were positively correlated with increased risperidone dosage in males(${\gamma}$=0.307 on the 2nd wks and ${\gamma}$=0.280 on the 4th wks), while they were not correlated with dosages in females. For the females, the PRL levels were negatively correlated(${\gamma}$=-0.320) with decrease of TSH concentration. The baseline PRL levels were not correlated with age, education, duration of illnesses, psychopathology, cognitive disorders in both males and females, while it was negatively correlated with TSH levels only in females(${\gamma}$=-0.320). 3) The cognitive dysfunction was not correlated with PRL levels in males, while PANSS-CF scores were negatively correlated with PRL levels(${\gamma}$=-0.220 on the 2nd wks and ${\gamma}$=-0.366 on the 4th wks) in females. The psychopathology was positively correlated with cognitive dysfunction in both males and females. Therefore, the risperidone-induced cognitive improvement seemed to be correlated with improvement of psychopathology in both males and females, and with increase in PRL levels only in females. Conclusions : The fact that the serum PRL levels of schizophrenics were higher than those of controls, especially in females suggested that it could be related with risperidone dosage in males and with primary pathological process in females. The risperidone-associated cognitive improvement seemed to be related with general improvement of psychopathology as well as the rise of PRL levels especially in females. The facts that the effect of risperidoneinduced hyperprolactinemia and the cognitive function were more in females suggested that somewhat different mechanisms could be exerted on them.