Public Perceptions and Barriers to Tuberculosis Treatment in Korea: A Large Language Model-Based Analysis of Naver Knowledge-iN Data from 2002 to 2024
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Objectives: This study was conducted to investigate public perceptions and concerns surrounding tuberculosis (TB) treatmentin Korea through an analysis of online queries about antitubercular medications. Additionally, it evaluated the effectivenessof large language models (LLMs) as analytical tools for processing unstructured healthcare data. Methods: Using LLMs,this study analyzed 44,174 questions that mentioned TB from Naver Knowledge-iN (2002–2024). Questions referencing antitubercularmedications were extracted and thematically categorized. Side effects were analyzed through parallel approachesexamining general and medication-specific effects. Questions about infectivity and social implications were further analyzedusing text embedding, dimensionality reduction, and clustering. The performance of LLMs was evaluated against human researchersand traditional methods. Results: Among questions mentioning specific medications (n = 919), rifampin (31.8%)and isoniazid (31.6%) were most frequently referenced. Of the 10,044 questions regarding antitubercular medication, managementchallenges represented the largest category (44.8%). Analysis of infectivity and social implications (n = 583) revealedpreviously unidentified concerns about blood donation and immigration eligibility. Employment-related concerns constitutedthe largest distinct subgroup (20.6%). Hepatotoxicity, dermatosis, and vomiting were the most frequently reported side effects. LLMs outperformed keyword matching in data processing and offered cost advantages over human analysis, with finetuningfurther reducing processing costs. Conclusions: This study produced novel insights into public concerns regarding TBtreatment and demonstrated the effectiveness of combining social media platform data with LLM-based analysis, providing asystematic framework for future healthcare research using unstructured public data and LLMs.
Abstract
Objectives: This study was conducted to investigate public perceptions and concerns surrounding tuberculosis (TB) treatmentin Korea through an analysis of online queries about antitubercular medications. Additionally, it evaluated the effectivenessof large language models (LLMs) as analytical tools for processing unstructured healthcare data. Methods: Using LLMs,this study analyzed 44,174 questions that mentioned TB from Naver Knowledge-iN (2002–2024). Questions referencing antitubercularmedications were extracted and thematically categorized. Side effects were analyzed through parallel approachesexamining general and medication-specific effects. Questions about infectivity and social implications were further analyzedusing text embedding, dimensionality reduction, and clustering. The performance of LLMs was evaluated against human researchersand traditional methods. Results: Among questions mentioning specific medications (n = 919), rifampin (31.8%)and isoniazid (31.6%) were most frequently referenced. Of the 10,044 questions regarding antitubercular medication, managementchallenges represented the largest category (44.8%). Analysis of infectivity and social implications (n = 583) revealedpreviously unidentified concerns about blood donation and immigration eligibility. Employment-related concerns constitutedthe largest distinct subgroup (20.6%). Hepatotoxicity, dermatosis, and vomiting were the most frequently reported side effects. LLMs outperformed keyword matching in data processing and offered cost advantages over human analysis, with finetuningfurther reducing processing costs. Conclusions: This study produced novel insights into public concerns regarding TBtreatment and demonstrated the effectiveness of combining social media platform data with LLM-based analysis, providing asystematic framework for future healthcare research using unstructured public data and LLMs.