A5101960398- Planetary Science and Exploration
- Astro and Planetary Science
- Topic Modeling
- Natural Language Processing Techniques
- Biomedical Text Mining and Ontologies
- Spacecraft Design and Technology
- Artificial Intelligence in Healthcare and Education
- Spacecraft and Cryogenic Technologies
- Gamma-ray bursts and supernovae
- Machine Learning in Healthcare
- Underwater Acoustics Research
- Space Science and Extraterrestrial Life
- Machine Learning in Bioinformatics
- Expert finding and Q&A systems
- Speech and Audio Processing
- Geological and Geochemical Analysis
- Computational Drug Discovery Methods
- Artificial Intelligence in Healthcare
- Spaceflight effects on biology
- Direction-of-Arrival Estimation Techniques
- Advanced Text Analysis Techniques
- Calcium signaling and nucleotide metabolism
- Cholangiocarcinoma and Gallbladder Cancer Studies
- Gastric Cancer Management and Outcomes
- Advanced Measurement and Detection Methods
Dankook University
2025
Korea University
2016-2024
Korea Maritime and Ocean University
2016-2018
University of Chicago
2010
In biomedical natural language processing, named entity recognition (NER) and normalization (NEN) are key tasks that enable the automatic extraction of entities (e.g. diseases drugs) from ever-growing literature. this article, we present BERN2 (Advanced Biomedical Entity Recognition Normalization), a tool improves previous neural network-based NER by employing multi-task model NEN models to achieve much faster more accurate inference. We hope our can help annotate large-scale texts for...
Abstract Motivation Traditional drug discovery approaches identify a target for disease and find compound that binds to the target. In this approach, structures of compounds are considered as most important features because it is assumed similar will bind same Therefore, structural analogs drugs selected candidates. However, even though not analogs, they may achieve desired response. A new method based on response, which can complement structure-based methods, needed. Results We implemented...
The recent outbreak of the novel coronavirus is wreaking havoc on world and researchers are struggling to effectively combat it. One reason why fight difficult due lack information knowledge. In this work, we outline our effort contribute shrinking knowledge vacuum by creating covidAsk, a question answering (QA) system that combines biomedical text mining QA techniques provide answers questions in real-time. Our also leverages retrieval (IR) approaches entity-level complementary models....
Molecular biomarkers that can predict drug efficacy in cancer patients are crucial components for the advancement of precision medicine. However, identifying these molecular remains a laborious and challenging task. Next-generation sequencing preclinical models have increasingly led to identification novel gene-mutation-drug relations, results been reported published scientific literature. Here, we present two new computational methods utilize all PubMed articles as domain specific...
Treating patients with major depressive disorder is challenging because it takes several months for antidepressants prescribed the to take effect. This limitation may result in increased risks and treatment costs. To address this limitation, an accurate antidepressant response prediction model needed. Recently, studies have proposed models that extract useful features such as neuroimaging biomarkers genetic variants from patient data, use them predictors predicting responses of patients....
Abstract Background Paclitaxel is commonly used as a second-line therapy for advanced gastric cancer (AGC). The decision to proceed with chemotherapy and select an appropriate regimen critical vulnerable patients AGC progressing after first-line chemotherapy. However, no predictive biomarkers exist identify who would benefit from paclitaxel-based Methods This study included 288 receiving between 2017 2022 part of the K-MASTER project, nationwide government-funded precision medicine...
Large Language Models (LLMs) have recently emerged, attracting considerable attention due to their ability generate highly natural, human-like text. This study compares the latent community structures of LLM-generated text and human-written within a hypothesis testing procedure. Specifically, we analyze three sets: original texts ($\mathcal{O}$), LLM-paraphrased versions ($\mathcal{G}$), twice-paraphrased set ($\mathcal{S}$) derived from $\mathcal{G}$. Our analysis addresses two key...
With the development of artificial intelligence (AI) technology centered on deep-learning, computer has evolved to a point where it can read given text and answer question based context text. Such specific task is known as machine comprehension. Existing comprehension tasks mostly use datasets general texts, such news articles or elementary school-level storybooks. However, no attempt been made determine whether an up-to-date deep learning-based model also process scientific literature...
We introduce HiPub, a seamless Chrome browser plug-in that automatically recognizes, annotates and translates biomedical entities from texts into networks for knowledge discovery. Using combination of two different named-entity recognition resources, HiPub can recognize genes, proteins, diseases, drugs, mutations cell lines in texts, achieve high precision recall. extracts entity-relationships to construct context-specific networks, integrates existing network data external databases It...
Abstract Objective Applications of machine learning in healthcare are high interest and have the potential to improve patient care. Yet, real-world accuracy these models clinical practice on different subpopulations remains unclear. To address important questions, we hosted a community challenge evaluate methods that predict outcomes. We focused prediction all-cause mortality as question. Materials Using Model-to-Data framework, 345 registered participants, coalescing into 25 independent...
Most electronic medical records, such as free-text radiological reports, are unstructured; however, the methodological approaches to analyzing these accumulating unstructured records limited. This article proposes a deep-transfer-learning-based natural language processing model that analyzes serial magnetic resonance imaging reports of rectal cancer patients and predicts their overall survival. To evaluate model, retrospective cohort study 4,338 was conducted. The experimental results...
To measure the ability of a machine to understand professional-level scientific articles, we construct question answering task called PaperQA. The PaperQA is based on more than 80 000 "fill-in-the-blank" type questions articles from reputed journals such as Nature and Science. We perform fine-grained linguistic analysis evaluation compare other conventional (QA) tasks general literature (e.g., books, news Wikipedia texts). results indicate that most difficult QA for both humans (lay people)...
The recent outbreak of the novel coronavirus is wreaking havoc on world and researchers are struggling to effectively combat it. One reason why fight difficult due lack information knowledge. In this work, we outline our effort contribute shrinking knowledge vacuum by creating covidAsk, a question answering (QA) system that combines biomedical text mining QA techniques provide answers questions in real-time. Our also leverages retrieval (IR) approaches entity-level complementary models....
Abstract Applications of machine learning in healthcare are high interest and have the potential to significantly improve patient care. Yet, real-world accuracy performance these models on different subpopulations remains unclear. To address important questions, we hosted a community challenge evaluate methods that predict outcomes. overcome privacy concerns, employed Model-to-Data approach, allowing citizen scientists researchers train private health data without direct access data. We...
In underwater acoustics, there has been many studies for finding the target direction using beamforming technique. When receiving a signal with frequency higher than design of array, it is difficult to estimate due spatial aliasing. this study, we propose method estimating array by frequency-wavenumber analysis. analysis performed, striation pattern appears, and confirmed that slope remains constant even if aliasing occurs. The was estimated visual inspection verified SAVEX15 data.
Many people seek majority opinions by searching for question-answers that are uploaded others or uploading their own questions on social media sites. However, have to read through a large number of documents returned search services find the opinions. Moreover, even when users upload sites, they cannot immediately obtain answers. To address these problems, we present Searching Majority Opinions System (SEMO), novel opinion-based system uses QA threads SNS and cQA websites. SEMO returns...
The Array Gain(AG) is a metric to assess the performance of an array and dependent on configuration array, frequency, as well directionality noise. In this study, AG calculated based spatial coherence between sensor elements in directional noise environment for given shape. estimated then compared with derived from sea going data signal ratio. results are presented discussed.