• Journal of Korea Water Resources Association
• /
• v.56 no.11
• /
• pp.823-835
• /
• 2023

Recently, due to the increase in abnormal climate, rainfall intensity is increasing and drought periods are continuing. These environmental changes lead to prolonged drought conditions and difficulties in real-time recognition. In general, drought can be judged by the amount of precipitation and the number of days without rainfall. In determining the impact of drought, it is divided into meteorological drought, agricultural drought, and hydrological drought and evaluation is made using the drought index, but environmental drought evaluation is insufficient. The river water quality managed through the total water pollution cap system is vulnerable to the effects of such drought. In this study, we aim to determine the drought impact on river water quality and quantify the impact of prolonged drought on water quality. The impact of rain-free days and accumulated precipitation on river water quality was quantitatively evaluated. The Load Duration Curve (LDC), which is used to evaluate the water quality of rivers, was used to evaluate water pollution occurring at specific times. It has been observed that when the number of consecutive rainless days exceeds 14 days, the target water quality in the mid-basin is exceeded in over 60% of cases. The cumulative rainfall is set at 28 days as the criteria, with an annual average rainfall of 3%, which is 32.1 mm or less. It has been noted that changes in water quality in rivers occur when there are 14 or more rainless days and the cumulative rainfall over 28 days is 32.1 mm or less in the Gamcheon Mid-basin. Based on the results of this study, it aims to quantify the drought impact and contribute to the development of a drought water quality index for future environmental droughts.

• Korean journal of applied entomology
• /
• v.62 no.4
• /
• pp.261-266
• /
• 2023

Interception data pertaining to Coleoptera, Hemiptera, Lepidoptera, Diptera, Thysanoptera, and Hymenoptera collected at the Korean quarantine border were cross-checked with new recorded species in Korea from 2015 to 2022. Overall, 45,084 interceptions belonging to the six orders were detected, and 545 species belonging to the six orders were newly reported in Korea. Of the 545 species, nine species were recorded as being intercepted at the quarantine border. Among the six orders, Coleoptera, Thysanoptera, and Hemiptera showed high numbers of interception; however, Hymenoptera revealed the highest number in new recorded species. These results indicate that recent newly recorded species in Korea are not subject to inspection at Korean borders and that the current quarantine system needs improvement.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.69-90
• /
• 2023

The purpose of this study is to evaluate the service attributes and consumer reactions of telemedicine apps in South Korea and visualize their differentiation by constructing positioning maps. We crawled 23,219 user reviews of 6 major telemedicine apps in Korea from the Google Play store. Topics were derived by BERTopic modeling, and sentiment scores for each topic were calculated through KoBERT sentiment analysis. As a result, five service characteristics in the application attribute category and three in the medical service category were derived. Based on this, a two-dimensional positioning map was constructed through principal component analysis. This study proposes an objective service evaluation method based on text mining, which has implications. In sum, this study combines empirical statistical methods and text mining techniques based on user review texts of telemedicine apps. It presents a system of service attribute elicitation, sentiment analysis, and product positioning. This can serve as an effective way to objectively diagnose the service quality and consumer responses of telemedicine applications.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.165-184
• /
• 2023

Recently, Deep learning analysis of unstructured text data using language models, such as Google's BERT and OpenAI's GPT has shown remarkable results in various applications. Most language models are used to learn generalized linguistic information from pre-training data and then update their weights for downstream tasks through a fine-tuning process. However, some concerns have been raised that privacy may be violated in the process of using these language models, i.e., data privacy may be violated when data owner provides large amounts of data to the model owner to perform fine-tuning of the language model. Conversely, when the model owner discloses the entire model to the data owner, the structure and weights of the model are disclosed, which may violate the privacy of the model. The concept of offsite tuning has been recently proposed to perform fine-tuning of language models while protecting privacy in such situations. But the study has a limitation that it does not provide a concrete way to apply the proposed methodology to text classification models. In this study, we propose a concrete method to apply offsite tuning with an additional classifier to protect the privacy of the model and data when performing multi-classification fine-tuning on Korean documents. To evaluate the performance of the proposed methodology, we conducted experiments on about 200,000 Korean documents from five major fields, ICT, electrical, electronic, mechanical, and medical, provided by AIHub, and found that the proposed plug-in model outperforms the zero-shot model and the offsite model in terms of classification accuracy.

• Phonetics and Speech Sciences
• /
• v.15 no.4
• /
• pp.71-80
• /
• 2023

This study primarily aimed to develop an automated stuttering identification and classification method using artificial intelligence technology. In particular, this study aimed to develop a deep learning-based identification model utilizing the convolutional neural networks (CNNs) algorithm for Korean speakers who stutter. To this aim, speech data were collected from 9 adults who stutter and 9 normally-fluent speakers. The data were automatically segmented at the phrasal level using Google Cloud speech-to-text (STT), and labels such as 'fluent', 'blockage', prolongation', and 'repetition' were assigned to them. Mel frequency cepstral coefficients (MFCCs) and the CNN-based classifier were also used for detecting and classifying each type of the stuttered disfluency. However, in the case of prolongation, five results were found and, therefore, excluded from the classifier model. Results showed that the accuracy of the CNN classifier was 0.96, and the F1-score for classification performance was as follows: 'fluent' 1.00, 'blockage' 0.67, and 'repetition' 0.74. Although the effectiveness of the automatic classification identifier was validated using CNNs to detect the stuttered disfluencies, the performance was found to be inadequate especially for the blockage and prolongation types. Consequently, the establishment of a big speech database for collecting data based on the types of stuttered disfluencies was identified as a necessary foundation for improving classification performance.

• Journal of the Korean Geotechnical Society
• /
• v.39 no.12
• /
• pp.47-60
• /
• 2023

At present, in South Korea, there is a growing concern regarding solar power facilities installed on slopes because they are prone to damage caused by natural disasters, such as heavy rainfall and typhoons. Each year, these solar power facilities experience soil erosion due to heavy rainfall and foundation damage or detachment caused by strong wind loads. Despite these challenges, the interaction between the ground and structures is not adequately considered. Current analyses primarily focus on the structural stability under external loads; the overall facility site's stability-excluding the solar structures-in relation to its surrounding slopes is neglected. Therefore, in this study, we use finite-difference method analysis to simulate the behavior of the foundation and piles to assess changes in lateral displacement and bending stress in piles, as well as the safety factor of sloped terrains, in response to various influencing factors, such as pile diameter, spacing between piles, pile-embedding depth, wind loads, and dry and wet conditions. The analysis results indicate that pile spacing and wind loads significantly influence lateral displacement and bending stress in piles, whereas pile-embedding depth strongly influences the safety factor of sloped terrains. Moreover, we found that under certain conditions, the design criteria in domestic standards may not be met.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.6
• /
• pp.526-532
• /
• 2023

Falls are fatal accidents that occur more than 420,000 times a year worldwide. Therefore, to study patients with falls, we found the association between extrinsic injury codes and principal diagnosis S-codes of patients with falls, and developed a prediction model to predict extrinsic injury codes based on the data of principal diagnosis S-codes of patients with falls. In this study, we received two years of data from 2020 and 2021 from Institution A, located in Gangneung City, Gangwon Special Self-Governing Province, and extracted only the data from W00 to W19 of the extrinsic injury codes related to falls, and developed a prediction model using W01, W10, W13, and W18 of the extrinsic injury codes of falls, which had enough principal diagnosis S-codes to develop a prediction model. 80% of the data were categorized as training data and 20% as testing data. The model was developed using MLP (Multi-Layer Perceptron) with 6 variables (gender, age, principal diagnosis S-code, surgery, hospitalization, and alcohol consumption) in the input layer, 2 hidden layers with 64 nodes, and an output layer with 4 nodes for W01, W10, W13, and W18 exogenous damage codes using the softmax activation function. As a result of the training, the first training had an accuracy of 31.2%, but the 30th training had an accuracy of 87.5%, which confirmed the association between the fall extrinsic code and the main diagnosis S code of the fall patient.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.18 no.3
• /
• pp.33-43
• /
• 2023

The paper employs a resource-based approach to analyze the relationship between institutional resources and faculty-led startup formation and performance in South Korean four-year universities from 2017 to 2021. The author proposes nine hypotheses to explain how institutional resources or systems affect the number of faculty startups, their employee numbers and the revenue of faculty-led startups, and compare four different groups of university resources for cross-college variation. The findings suggest that institutional factors impacting faculty-led startup performance differ from those impacting other categories of startups. Universities should provide a more favorable environment, including flexible personnel policies and accompanying startup support infrastructure, to encourage faculty-led startups. In contrast, it is more effective for better performance of faculty startups, in terms of their job creation and revenue, to have more financial resources and good paper publications. The results also suggest that university technology-holding companies are crucial for increasing the number of professor startups and their performance. These findings have implications for both university and government policymakers, who aim to facilitate greater participation of professors in startup formation and commercialization of technology.

• Pediatric Infection and Vaccine
• /
• v.31 no.1
• /
• pp.46-54
• /
• 2024

Purpose: This study aimed to identify the pathogens of bloodstream infection in children with underlying hemato-oncologic diseases, analyze susceptibility patterns, compare temporal trends with those of previous studies, and assess empirical antimicrobial therapy. Methods: Retrospective review study of children bacteremia in hemato-oncologic diseases was conducted at Seoul National University Bundang Hospital from January 2013 to July 2023. Results: Overall, 98 episodes of bacteremia were observed in 74 patients. Among pathogens isolated, 57.1% (n=56) were Gram-positive bacteria, 38.8% (n=38) were Gram-negative bacteria, and 4.1% (n=4) were Candida spp. The most common Gram-positive bacteria were coagulase-negative staphylococci (n=21, 21.4%) and Staphylococcus aureus, (n=14, 14.3%) whereas the most common Gram-negative bacteria were Klebsiella pneumoniae (n=16, 16.3%) and Escherichia coli (n=10, 10.2%). The susceptibility of Gram-positive bacteria to penicillin, oxacillin, and vancomycin was 11.5%, 32.7%, and 94.2%, respectively and the susceptibility of Gram-negative bacteria to cefotaxime, piperacillin/tazobactam, imipenem, gentamicin, and amikacin was 68.6%, 80%, 97.1%, 82.9%, and 91.4%, respectively. Methicillin-resistant S. aureus was detected in 1 strain and among Gram-negative strains, extended spectrum β-lactamase accounted for 28.9% (12/38). When analyzing the antibiotic susceptibility and empirical antibiotics, the mismatch rate was 25.5% (n=25). The mortality rate of children within 30 days of bacteremia was 7.1% (n=7). Conclusions: Empirical antibiotic therapy for bacteremia in children with hemato-oncologic diseases should be based on the local antibiogram in each institution and continuous monitoring is necessary.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.6
• /
• pp.260-268
• /
• 2024

Cervical cancer is the fourth most common cancer in women worldwide, and more than 604,000 new cases were reported in 2020 alone, resulting in approximately 341,831 deaths. The Cox regression model is a major model widely adopted in cancer research, but considering the existence of nonlinear associations, it faces limitations due to linear assumptions. To address this problem, this paper proposes ResSurvNet, a new model that improves the accuracy of cervical cancer mortality prediction using ResNet's residual learning framework. This model showed accuracy that outperforms the DNN, CPH, CoxLasso, Cox Gradient Boost, and RSF models compared in this study. As this model showed accuracy that outperformed the DNN, CPH, CoxLasso, Cox Gradient Boost, and RSF models compared in this study, this excellent predictive performance demonstrates great value in early diagnosis and treatment strategy establishment in the management of cervical cancer patients and represents significant progress in the field of survival analysis.