Young-Kyoon Suh

Prof. Young-Kyoon Suh

Kyungpook National University
orcid

Research Focus & Metrics

Academic Career Length

20Years (2007-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
31 January 2023 in Journal of Medical Internet Research
Verified
Journal Article
Open access
SPSusan Park
Susan Park
YSYoung-Kyoon Suh
Young-Kyoon Suh
Background: The unprecedented speed of COVID-19 vaccine development and approval has raised public concern about its safety. However, studies on public discourses and opinions on social media focusing on adverse events (AEs) related to COVID-19 vaccine are rare. Objective: This study aimed to analyze Korean tweets about COVID-19 vaccines (Pfizer, Moderna, AstraZeneca, Janssen, and Novavax) after the vaccine rollout, explore the topics and sentiments of tweets regarding COVID-19 vaccines, and examine their changes over time. We also analyzed topics and sentiments focused on AEs related to vaccination using only tweets with terms about AEs. Methods: We devised a sophisticated methodology consisting of 5 steps: keyword search on Twitter, data collection, data preprocessing, data analysis, and result visualization. We used the Twitter Representational State Transfer application programming interface for data collection. A total of 1,659,158 tweets were collected from February 1, 2021, to March 31, 2022. Finally, 165,984 data points were analyzed after excluding retweets, news, official announcements, advertisements, duplicates, and tweets with <2 words. We applied a variety of preprocessing techniques that are suitable for the Korean language. We ran a suite of analyses using various Python packages, such as latent Dirichlet allocation, hierarchical latent Dirichlet allocation, and sentiment analysis. Results: The topics related to COVID-19 vaccines have a very large spectrum, including vaccine-related AEs, emotional reactions to vaccination, vaccine development and supply, and government vaccination policies. Among them, the top major topic was AEs related to COVID-19 vaccination. The AEs ranged from the adverse reactions listed in the safety profile (eg, myalgia, fever, fatigue, injection site pain, myocarditis or pericarditis, and thrombosis) to unlisted reactions (eg, irregular menstruation, changes in appetite and sleep, leukemia, and deaths). Our results showed a notable difference in the topics for each vaccine brand. The topics pertaining to the Pfizer vaccine mainly mentioned AEs. Negative public opinion has prevailed since the early stages of vaccination. In the sentiment analysis based on vaccine brand, the topics related to the Pfizer vaccine expressed the strongest negative sentiment. Conclusions: Considering the discrepancy between academic evidence and public opinions related to COVID-19 vaccination, the government should provide accurate information and education. Furthermore, our study suggests the need for management to correct the misinformation related to vaccine-related AEs, especially those affecting negative sentiments. This study provides valuable insights into the public discourses and opinions regarding COVID-19 vaccination.

27 April 2021 in Scientific Programming
Verified
Journal Article
Open access
PDF file
JOJin-Sung Ok
Jin-Sung Ok
SKSoon-Do Kwon
Soon-Do Kwon
CHCheol-Eun Heo
Cheol-Eun Heo
YSYoung-Kyoon Suh
Young-Kyoon Suh
The development of industrial Internet of Things (IIoT), big data, and artificial intelligence technologies is leading to a major change in the production system. The change is being propagated into the wave of transforming the existing system with a vertical structure into the corresponding horizontal platform or middleware. Accordingly, the way of acquiring IIoT data from an individual system is being altered to the way of being increasingly centralized through an integrated middleware of a scalable server or through a large platform. That said, middleware-based IIoT data acquisition must consider multiple factors, such as infrastructure (e.g., operation environment and network), protocol heterogeneity, interoperability (e.g., links with legacy systems), real-time, and security. This manuscript explains these five aspects in detail and provides a taxonomy of eighteen state-of-the-art IIoT data-acquisition middleware systems based on these aspects. To validate one of these aspects (network), we present our evaluation results at a real production site where IIoT data-acquisition loss rates are compared between wireless (long-term evolution) and wired networks. As a result, the wired communication can be more suitable for centralized IIoT data-acquisition middleware than wireless networks. Finally, we discuss several challenges in establishing the best IIoT data-acquisition middleware in a centralized way.

18 July 2019 in IEEE Access
Verified
Journal Article
Open access
SKSeounghyeon Kim
Seounghyeon Kim
YSYoung-Kyoon Suh
Young-Kyoon Suh
JKJeeyoung Kim
Jeeyoung Kim
In recent years, computational science and engineering (CSE) simulations using high-performance computing resources are actively exploited to solve complex domain-specific problems. Thanks to the remarkable advance of IT technology, the CSE community is challenging more complex and difficult problems than ever before, by running these simulations online. In this regard, we often witness that 1) online simulation users suffer from knowing little about the estimated termination time of their launched simulations and 2) the limited computing resources are squandered by wrong input that leads the simulations to run forever. To address such issues, we propose a novel execution time estimation scheme, termed EXTES, using machine learning techniques for more efficient online CSE simulations. With a large amount of existing provenance data, the EXTES scheme trains a suite of models rooted from classification, regression, and a hybrid of the two and utilize these models to estimate the execution time for specified input parameters for simulations. In the experiments on real simulation data, our proposed models achieved about 73% accuracy on average in execution time estimation across 16 simulation programs taken from a variety of CSE fields. In the meantime, the overhead incurred by the training and estimation is almost negligible.

4 December 2020 in IEEE Access
Verified
Journal Article
Open access
YSYoung-Kyoon Suh
Young-Kyoon Suh
SKSeounghyeon Kim
Seounghyeon Kim
JKJeeyoung Kim
Jeeyoung Kim
Efficient scheduling among simultaneous simulation jobs is of critical importance in the allocation of limited computing and I/O resources. The difficulty of predicting when a job is completed can cause nontrivial problems for system administrators and users e.g., squandered resources, long waiting times, and simulation plan delays. To alleviate these problems, we propose a novel simulation runtime estimation scheme termed CLUTCH, which employs a well-orchestrated ensemble of clustering, classification, and regression techniques. The proposed scheme trains a runtime estimation model through a series of steps: (i) grouping past simulation provenance records by clustering, (ii) labeling each of the grouped records by classification, and (iii) performing regression on the execution times in each group. Given a simulation and its external arguments, the trained model predicts the simulation’s runtime with high accuracy in a black box fashion, using only basic external arguments without needing extra information. We additionally propose two optimization algorithms which significantly reduce training overhead without sacrificing estimation quality. In the experiment with real datasets, our model achieved approximately a 14.2% growth in estimation accuracy, compared to the most recent state-of-the-art method; with our optimizations applied, the model was trained 16 times faster while still retaining accuracy.

20 September 2021 in The VLDB Journal
Verified
Journal Article
Closed access
PDF file
RSRichard T. Snodgrass
Richard T. Snodgrass
SCSabah Currim
Sabah Currim
YSYoung-Kyoon Suh
Young-Kyoon Suh

8 March 2023 in Journal of Medical Internet Research
Verified
Journal Article
Open access
KKKarina Karapetian
Karina Karapetian
SJSoo Min Jeon
Soo Min Jeon
JKJinwon Kwon
Jinwon Kwon
YSYoung-Kyoon Suh
Young-Kyoon Suh
Background: Drug-induced suicide has been debated as a crucial issue in both clinical and public health research. Published research articles contain valuable data on the drugs associated with suicidal adverse events. An automated process that extracts such information and rapidly detects drugs related to suicide risk is essential but has not been well established. Moreover, few data sets are available for training and validating classification models on drug-induced suicide. Objective: This study aimed to build a corpus of drug-suicide relations containing annotated entities for drugs, suicidal adverse events, and their relations. To confirm the effectiveness of the drug-suicide relation corpus, we evaluated the performance of a relation classification model using the corpus in conjunction with various embeddings. Methods: We collected the abstracts and titles of research articles associated with drugs and suicide from PubMed and manually annotated them along with their relations at the sentence level (adverse drug events, treatment, suicide means, or miscellaneous). To reduce the manual annotation effort, we preliminarily selected sentences with a pretrained zero-shot classifier or sentences containing only drug and suicide keywords. We trained a relation classification model using various Bidirectional Encoder Representations from Transformer embeddings with the proposed corpus. We then compared the performances of the model with different Bidirectional Encoder Representations from Transformer–based embeddings and selected the most suitable embedding for our corpus. Results: Our corpus comprised 11,894 sentences extracted from the titles and abstracts of the PubMed research articles. Each sentence was annotated with drug and suicide entities and the relationship between these 2 entities (adverse drug events, treatment, means, and miscellaneous). All of the tested relation classification models that were fine-tuned on the corpus accurately detected sentences of suicidal adverse events regardless of their pretrained type and data set properties. Conclusions: To our knowledge, this is the first and most extensive corpus of drug-suicide relations.