2026-03-01 — Paper Published in Scientific Data: "Telehealth Infrastructure for Cancer Care in the United States." Learn more →
2026-01-15 — Paper Published in IJGIS: "Evaluating the Impact of Open Data + FAIR Policies on Computational Reproducibility: A Systematic Analysis of IJGIS Publications" (with Peter Kedron). Learn more →
2026-10-02/03 — CGA Conference 2026, Harvard CGA, Cambridge MA. Learn more →
2026-03-20 — Presentation at AAG 2026: “Graph Conditional Regression”. Learn more →
Geospatial Analytics Extension for KNIME, v 2.0 — An open-source visual programming platform for geospatial analysis, supporting both local and cloud environments, and used by over 200,000 users (by 2025-4-12). Recognized as the top community extension for KNIME, it led to Lingbo's nomination as KNIME's Contributor of the Month for March 2025. Textbook | GitHub | KNIME Hub workflows | Datasets on Harvard Dataverse
XGeoML, v0.6 — The first Python package for ensemble explainable geospatial machine learning, integrating over 30 machine learning models with local geospatial modelling; downloaded over 15,000 times on PYPI (by 2025-8-4). Download statistics →
MIGIS is an interdisciplinary research initiative that integrates spatial thinking, geospatial modeling, and explainable artificial intelligence to address critical challenges in urban health, environmental justice, and data-driven decision-making. Led by Dr. Lingbo Liu at Harvard University’s Center for Geographic Analysis, MIGIS emphasizes not only the technical rigor of geospatial intelligence but also the ethical, human-centered, and policy-relevant dimensions of spatial data science. MIGIS stands at the intersection of spatial modeling, digital equity, and open science. Its core mission is to develop mindful and responsible spatial analytics—tools and methods that are not only powerful but also interpretable, inclusive, and actionable. The initiative contributes to global research through open-source toolkits, reproducible workflows, and collaborative platforms that empower governments, researchers, and communities to build healthier and more equitable cities.
Explainable geospatial machine learning models for public health and climate resilience.
Learn more →
Accessibility modeling and urban health analysis based on human mobility and spatial equity.
Learn more →
KNIME-based visual tools for geospatial workflows, teaching, and reproducible research.
Explore datasets →
An open, reusable ZCTA-level dataset integrating oncologist capacity, broadband and 5G access, affordability, and road travel times to measure in-person and telehealth accessibility with 2SFCA/2SVCA.
Learn more →
A multimodal framework that fuses OpenStreetMap text and satellite imagery on H3 grids to classify semantics from high-frequency GPS trajectories for mental-health geomatics.
Learn more →
National 2SFCA/2SVCA modeling across 33,499 ZCTAs shows telehealth can reduce but not remove rurality-deprivation inequities because digital infrastructure remains uneven.
Learn more →
Introduces r2SFCA, an entropy-based calibration approach that aligns demand-side accessibility and supply-side crowdedness by optimizing a unified distance-decay function.
Learn more →