텍스트 회귀분석을 이용한 수소에너지 R&D 프로젝트 연구비 예측

초록

In response to the imperative of mitigating greenhouse gas emissions to address climate change, the global adoption of renewable energy sources has been significantly accelerated, with hydrogen emerging as a promising alternative to fossil fuels due to its high energy density and carbon-neutral characteristics. Consequently, major industrialized nations are committing substantial financial resources to hydrogen-related research and development while implementing comprehensive policy support frameworks. South Korea has similarly prioritized strategic research and development initiatives, particularly focusing on hydrogen fuel cell vehicles and fuel cell technologies. This study develops a predictive model for government research investment and total research funding by applying text regression analysis to 3,540 hydrogen R&D project technical documents obtained from the National Science and Technology Information Service (NTIS). The empirical analysis demonstrates that the Lasso regression model achieves optimal predictive performance, with keywords including "technology," "demonstration," and "hydrogen charging station" identified as critical predictive variables. This research contributes to methodological diversity in academic literature by applying text regression analysis to the hydrogen R&D domain, while simultaneously providing practical implications for evidence-based policy formulation through budget prediction capabilities.

키워드