Topic Identification and Spatiotemporal Diffusion Analysis on Hydrogen Energy Technology News
Han Guoyong
Gao Yang
Wu Yixuan
Yan Wanxuan
Wu Jinxin
Abstract:Objective This study aims to reveal the release,diffusion and implementation of policies,reflect the de-velopmental stages of related technology application,and provide decision-making support for policy formulation and industrial development through analysis on hydrogen energy news texts.Methods Firstly,representative hydro-gen energy news texts from China Energy News were collected as the dataset,and an LDA model was employed to identify core topics.Secondly,an indicator system comprising three dimensions-diffusion intensity,breadth and dis-tribution was introduced to study the news diffusion status in multiple dimensions.Finally,an in-depth analysis was conducted from aspects of data distribution,topic distribution,temporal evolution,spatial distribution similarity and spatial evolution.Results The regional development of hydrogen energy in our country is unbalanced,mainly influ-enced by factors such as central and local policies,research institutions,geographical location and resources.Chi-na's hydrogen energy industry has formed a development path of"policy guidance-technological breakthroughs-regional pilots",presenting a compound development pattern of"multi-dimensional gradient evolution+networked coordination".However,further breakthroughs are still needed in aspects such as cross-regional coordination,bal-anced development of technology chains and construction of market-oriented mechanisms to achieve the key leap from demonstration application to large-scale development.Approach In the future,semantic-related models will be utilized to conduct in-depth analysis on papers and patent data related to hydrogen energy,exploring its specific technological evolution characteristics and development trends.
Keywords:Hydrogen energy newsTopic identificationTopic diffusionSpatial similaritySpatiotemporal evolution
Publication Date:2025-10-30
Online Publishing Date:2025-11-27(First online date of this platform, not the publication date of the document)
Pages:14( 15-27,37 )
