A5026857815- Topic Modeling
- Data-Driven Disease Surveillance
- Sentiment Analysis and Opinion Mining
- COVID-19 epidemiological studies
- Lung Cancer Treatments and Mutations
- COVID-19 diagnosis using AI
- Misinformation and Its Impacts
- Artificial Intelligence in Healthcare
- Multimodal Machine Learning Applications
- Influenza Virus Research Studies
- Online Learning and Analytics
- Colorectal Cancer Treatments and Studies
- Natural Language Processing Techniques
- Machine Learning in Healthcare
- Gastric Cancer Management and Outcomes
- Urinary and Genital Oncology Studies
- Generative Adversarial Networks and Image Synthesis
- Lung Cancer Research Studies
- Neural Networks and Applications
- Brain Metastases and Treatment
Catholic University of Korea
2019-2025
Seoul St. Mary's Hospital
2012
In the field of natural language processing (NLP), prompt-based learning is widely used for efficient parameter learning. However, this method has drawback shortening input length by extent attached prompt, leading to inefficient utilization space. study, we propose P-Distill, a novel prompt compression that mitigates aforementioned limitation while maintaining performance via knowledge distillation. The distillation process P-Distill consists two methods, namely initialization and...
Predicting student dropout from universities is an imperative but challenging task. Numerous data-driven approaches that utilize both demographic information (e.g., gender, nationality, and high school graduation year) academic GPA, participation in activities, course evaluations) have shown meaningful results. Recently, pretrained language models achieved very successful results understanding the tasks associated with structured data as well textual data. In this paper, we propose a novel...
Political perspective detection in news media—identifying political bias articles—is an essential but challenging low-resource task. Prompt-based learning (i.e., discrete prompting and prompt tuning) achieves promising results scenarios by adapting a pre-trained model to handle new tasks. However, these approaches suffer performance degradation when the target task involves textual domain (e.g., domain) different from pre-training masked language modeling on general corpus). In this paper,...
As microblogs have become commonplace, recommending relevant hashtags for microblog posts has increasingly important. However, appropriate a post is challenging because it requires high-level understanding of the context and relationships information in post. In this paper, we propose novel hashtag recommendation framework that incorporates external knowledge to enrich posts. Using an image post, obtain hierarchical extracted by Open Directory Project (ODP)-based classifier. Experimental...
<sec> <title>BACKGROUND</title> The number of confirmed coronavirus disease (COVID-19) cases is a crucial indicator policies and lifestyles. Previous studies have attempted to forecast using machine learning techniques that utilize previous case counts search engine queries predetermined by experts. However, they limitations in reflecting temporal variations associated with pandemic dynamics. </sec> <title>OBJECTIVE</title> We propose novel framework extract keywords highly COVID-19,...
The number of confirmed COVID-19 cases is a crucial indicator policies and lifestyles. Previous studies have attempted to forecast using machine learning techniques that use previous case counts search engine queries predetermined by experts. However, they limitations in reflecting temporal variations associated with pandemic dynamics.