Global News Corpus
Approximately 200 million multilingual news articles over five years, stored in PostgreSQL with automated daily ingestion.
AI research infrastructure for geoscience, sustainability, and human well-being
Visiting Assistant Professor, Kyushu University | Chief Data Scientist & Head of AI Research, aiESG Inc.
I build large-scale research data systems and AI models that connect global news, remote sensing, climate projections, ESG text, and well-being microdata.
Current profile
My work sits at the intersection of artificial intelligence, geoscience, sustainability economics, and social well-being. Over the last three years I have focused on building research-ready data infrastructure at global scale, then using it to produce empirical AI and sustainability research.
Research infrastructure
Approximately 200 million multilingual news articles over five years, stored in PostgreSQL with automated daily ingestion.
Gallup World Poll, Global Flourishing Study, and DHS alignment at tens-of-millions respondent scale through ADB collaboration.
MODIS, CMIP6, population, land cover, air pollution, and greenness grids for long-term climate and demographic projection.
A prototype corporate social-risk database built from global news, with stronger entity resolution and scoring planned.
Gemini-based agents for cross-survey variable alignment, with literature screening and robustness-checking agents on the roadmap.
Dedicated PostgreSQL data server, Google Cloud Platform compute, and access to shared 4xH100 GPU nodes for large-model experiments.
Current research directions
Global monthly 5-km forecasts for 2000-2100 using MODIS and CMIP6, with 500-m downscaling and vegetation-index transfer learning planned.
Cross-survey harmonization for research on gender, greenness, pollution, loneliness, rural-urban disparity, and parental relationships.
Event-level modeling of news impacts on stock prices, investor sentiment, ESG tone, and corporate social-risk signals.
Open-source software
I maintain GWPR.light on CRAN and publish open-source tools for spatial statistics, AI-assisted academic submission workflows, and personal schedule planning.
The first R package for geographically weighted panel regression, reused across my own environmental and well-being studies.
install.packages("GWPR.light")
CRAN
A web-based tool for generating academic manuscript submission documents, including cover letters, title pages, highlights, CRediT statements, and declarations.
python STAGE/app.py
GitHub
A personal schedule management assistant for adding tasks through natural language or voice, arranging them on a timeline, and using Gemini AI to plan the day.
./Zhixiang/init.sh
GitHub
Selected publications
Business Strategy and the Environment.
2026Strategic Business Research, 2, 100237.
2026Humanities and Social Sciences Communications, 13, 601.
2026Sustainable Cities and Society, 141, 107273.
2026Journal of Cleaner Production, 552, 148017.
Publication records
Contact