<link>https://ankky-zhao.github.io/</link></image><item><title>Web Scraping Roadmap</h1> <article> <h1>Web Scraping Roadmap</h1> <p>Mon, 19 Oct 2026 00:00:00 +0000</p> <p>The roadmap started as a guide for a friend learning to collect web data and grew into a public resource. It covers requests and parsing, dynamic pages, data cleaning and storage, and good practice around rate limits and site terms, with every stage shown both in hand-written code and with AI coding assistants.</p> </article> <article> <h1>LLM Agent Roadmap</h1> <p>Tue, 15 Sep 2026 00:00:00 +0000</p> <p>The roadmap organises the field into stages, from foundations and tool use to retrieval, multi-agent systems, agentic reinforcement learning and safety. Each module links to curated tutorials, courses and code, and a timeline traces key papers from 2020 onwards.</p> <p>Content lives in JSON files separate from the page, so others can propose resources through pull requests. The site is refreshed on a monthly cycle.</p> </article> <article> <h1>Geo-CLASS: Knowledge from Earth Engine Scripts</h1> <p>Mon, 03 Nov 2025 00:00:00 +0000</p> <p>Thousands of Google Earth Engine scripts shared online encode practical knowledge about how geospatial analyses are built, from data selection and cloud masking to index calculation and export. This knowledge is scattered across code and hard to reuse.</p> <p> <figure > <div class="flex justify-center "> <div class="w-full" > </div> </div></figure> <em>Extracting themes, functions, data sources and spatial and temporal scopes from a script into a knowledge graph.</em></p> <p>Geo-CLASS coordinates a large language model with small language models. Guided by a domain schema and a schema-aware, chain-of-thought prompt, the LLM recognises entities and relations in each script. The SLMs then map the free-text descriptions onto a geoscience knowledge base (GCMD, GeoNames, the GEE data catalogue and a functional dictionary) through semantic vectorisation, cosine similarity and best-match selection, so that synonymous descriptions from different scripts end up as the same standardised entity.</p> <p>On 237 scripts sampled from 295,943, the framework improved entity and relation recognition accuracy by up to 31.9% and 12.0% over LLM baselines including Llama-3, GPT-3.5 and GPT-4o.</p> <p>The overview figure at the top shows the full Geo-CLASS framework. Figures from Zhao et al. (2026), <em>International Journal of Geographical Information Science</em>, 40(3), 808–841, .</p> </article> <article> <h1>Collaborate</h1> <p>Tue, 20 Oct 2026 00:00:00 +0000</p> </article> <article> <h1>News</h1> <p>Tue, 20 Oct 2026 00:00:00 +0000</p> </article> <article> <h1>Profile</h1> <p>Tue, 20 Oct 2026 00:00:00 +0000</p> </article> <article> <h1>Why I built an open learning map for LLM agents</h1> <p>Tue, 15 Sep 2026 00:00:00 +0000</p> <p>When I started working on agents for geospatial AI, most of my time went into finding out what to read next. Blog posts, courses and papers were plentiful, but the order in which they made sense was not obvious, and many resources were only available in one language.</p> <p>The is my attempt to write that order down. It groups the field into stages, links each stage to resources I found useful, and keeps a timeline of papers that shifted the direction of the field. Keeping the content in plain JSON makes it easy for others to correct or extend it, and a monthly update keeps it from going stale.</p> <p>It is a personal map rather than a complete survey, so suggestions are very welcome on .</p> </article> <article> <h1>Distributed Clustering by Local Direction Centrality on Spark</h1> <p>Sun, 01 Jun 2025 00:00:00 +0000</p> <p>Parallelises the clustering-by-local-direction-centrality algorithm on Apache Spark with spatial partitioning and performance optimisations for large datasets.</p> <p><strong>Citation</strong><br> Gui Z., Huang Z., Chen H., Peng D., Liu Y., Fang G., Lan Q., <strong>Zhao A.</strong>, Wu H. (2025). A performance-enhanced distributed computing framework for clustering by local direction centrality upon Apache Spark. Big Earth Data, 9(4), 797–827. </p> <p><em>Cover figure: Enhanced framework of the distributed D-CDC algorithm. From Gui et al. (2025), Big Earth Data.</em></p> </article> <article> <h1>Dynamic Visual Sub-Embeddings for Image-Text Retrieval</h1> <p>Sun, 01 Jun 2025 00:00:00 +0000</p> <p>Learns several dynamic visual sub-embeddings per image to capture varied semantics and adds a fast re-ranking stage to improve image-text retrieval.</p> <p><strong>Citation</strong><br> Wei W., Gui Z., Wu C., <strong>Zhao A.</strong>, Peng D., Wu H. (2025). Dynamic visual semantic sub-embeddings and fast re-ranking for image-text retrieval. IEEE Transactions on Multimedia, 27, 3781–3796. </p> <p><em>Cover figure: Overview of the model. From Wei et al. (2025), IEEE Transactions on Multimedia.</em></p> </article> <article> <h1>GEE-OPs: Operator Knowledge Base for Earth Engine Code</h1> <p>Sun, 01 Jun 2025 00:00:00 +0000</p> <p>Mines operator call relationships from Earth Engine scripts with syntax-tree parsing and frequent-pattern analysis, forming a knowledge base that improves LLM code generation.</p> <p><strong>Citation</strong><br> Hou S., Liang J., <strong>Zhao A.</strong>, Wu H. (2025). GEE-OPs: an operator knowledge base for geospatial code generation on Google Earth Engine powered by large language models. Geo-spatial Information Science. </p> <p><em>Cover figure: Framework for constructing the GEE-OPs knowledge base. From Hou et al. (2025), Geo-spatial Information Science.</em></p> </article> <article> <h1>Geo-FuB: Operator-Function Knowledge Base for Geospatial Code</h1> <p>Sun, 01 Jun 2025 00:00:00 +0000</p> <p>Builds an operator-function knowledge base from Earth Engine scripts, linking operators to the functions they serve, and uses it as retrieval context for geospatial code generation with LLMs.</p> <p><strong>Citation</strong><br> Hou S., <strong>Zhao A.</strong>, Liang J., Shen Z., Wu H. (2025). Geo-FuB: constructing an operator-function knowledge base for geospatial code generation with large language models. Knowledge-Based Systems, 319, 113624. </p> <p><em>Cover figure: Workflow of the Geo-FuB framework construction. From Hou et al. (2025), Knowledge-Based Systems.</em></p> </article> <article> <h1>GeoCode-GPT: A Large Language Model for Geospatial Code</h1> <p>Sun, 01 Jun 2025 00:00:00 +0000</p> <p>Fine-tunes an open LLM on purpose-built geospatial code corpora for pretraining, instruction tuning and evaluation, giving a model specialised in generating geospatial code.</p> <p><strong>Citation</strong><br> Hou S., Shen Z., <strong>Zhao A.</strong>, Liang J., Gui Z., Guan X., Li R., Wu H. (2025). GeoCode-GPT: a large language model for geospatial code generation. International Journal of Applied Earth Observation and Geoinformation, 104456. </p> <p><em>Cover figure: Technical workflow of GeoCode-GPT. From Hou et al. (2025), International Journal of Applied Earth Observation and Geoinformation.</em></p> </article> <article> <h1>GeoCogent: An LLM Agent for Geospatial Code Generation</h1> <p>Sun, 01 Jun 2025 00:00:00 +0000</p> <p>An LLM-based agent that plans, retrieves geospatial knowledge, and generates and iteratively debugs code to complete geospatial programming tasks.</p> <p><strong>Citation</strong><br> Hou S., Jiao H., Liang J., Shen Z., <strong>Zhao A.</strong>, Wu H. (2025). GeoCogent: an LLM-based agent for geospatial code generation. International Journal of Geographical Information Science. </p> <p><em>Cover figure: Overall architecture of GeoCogent. From Hou et al. (2025), International Journal of Geographical Information Science.</em></p> </article> <article> <h1>GeoColab: Multi-Agent Collaboration for Geospatial Code</h1> <p>Sun, 01 Jun 2025 00:00:00 +0000</p> <p>Coordinates role-specialised LLM agents in a collaborative workflow that decomposes, writes, reviews and refines geospatial code.</p> <p><strong>Citation</strong><br> Wu H., Jiao H., Hou S., Liang J., Shen Z., <strong>Zhao A.</strong>, et al. (2025). GeoColab: an LLM-based multi-agent collaborative framework for geospatial code generation. International Journal of Digital Earth, 18(2). </p> <p><em>Cover figure: Concept of the GeoColab multi-agent system. From Wu et al. (2025), International Journal of Digital Earth.</em></p> </article> <article> <h1>Semantic-Driven Geospatial Modeling Knowledge Graph</h1> <p>Sun, 01 Jun 2025 00:00:00 +0000</p> <p>Designs a geospatial modeling knowledge graph whose entities and relations are extracted with LLMs, supporting semantic retrieval and recommendation of modeling knowledge.</p> <p><strong>Citation</strong><br> Liang J., Hou S., <strong>Zhao A.</strong>, Xu Q., Xiang L., Li R., Wu H. (2025). Design and application of a semantic-driven geospatial modeling knowledge graph based on large language models. Geo-spatial Information Science. </p> <p><em>Cover figure: Overview of the co-augmentation framework. From Liang et al. (2025), Geo-spatial Information Science.</em></p> </article> <article> <h1>Trajectory Traits: Profiling Drivers from GPS Trajectories</h1> <p>Sun, 01 Jun 2025 00:00:00 +0000</p> <p>Introduces a Trajectory Trait Scale inspired by the five-factor personality model, scoring drivers’ extroversion, openness, neuroticism and conscientiousness from GPS trajectory features.</p> <p><strong>Citation</strong><br> Liu Y., Gui Z., Xu Y., Gao S., <strong>Zhao A.</strong>, et al. (2025). Profiling mobility patterns and driving behaviors of individual drivers via trajectory trait. The Innovation Geoscience, 3(1), 100114. </p> <p><em>Graphical abstract from Liu et al. (2025), The Innovation Geoscience.</em></p> </article> <article> <h1>GPT-Enhanced Knowledge Extraction for Earth Engine Models</h1> <p>Sat, 01 Jun 2024 00:00:00 +0000</p> <p>A GPT-assisted workflow that extracts structured knowledge about geographic analysis models from Google Earth Engine scripts and reuses it to support model construction.</p> <p><strong>Citation</strong><br> Liang J., <strong>Zhao A.</strong>, Hou S., Wu H., Jin F. (2024). A GPT-enhanced framework on knowledge extraction and reuse for geographic analysis models in Google Earth Engine. International Journal of Digital Earth, 17(1). </p> <p><em>Cover figure: Prompt design and interaction with ChatGPT. From Liang et al. (2024), International Journal of Digital Earth.</em></p> </article> <article> <h1>Map Retrieval Intention Recognition with Geographic Semantics</h1> <p>Sat, 01 Jun 2024 00:00:00 +0000</p> <p>Recognises users’ map retrieval intentions from relevance feedback, with geographic ontologies guiding intention inference and its iterative updates.</p> <p><strong>Citation</strong><br> Gui Z., Liu X., <strong>Zhao A.</strong>, et al. (2024). Map retrieval intention recognition based on relevance feedback and geographic semantic guidance. Information Processing & Management, 61(4), 103767. </p> <p><em>Cover figure: Map retrieval workflow enhanced by intention recognition. From Gui et al. (2024), Information Processing & Management.</em></p> </article> <article> <h1>Radiometric Normalisation of Cross-Sensor Gaofen Images</h1> <p>Tue, 01 Jun 2021 00:00:00 +0000</p> <p>Normalises radiometry across Gaofen sensors by detecting unchanged pixels with change detection and a chi-square test, then fitting a relative normalisation model.</p> <p><strong>Citation</strong><br> Yan L., Yang J., Zhang Y., <strong>Zhao A.</strong>, Li X. (2021). Radiometric normalization for cross-sensor optical Gaofen images with change detection and chi-square test. Remote Sensing, 13(16), 3125. </p> <p><em>Cover figure: Bitemporal change detection stage of the proposed relative radiometric normalisation method (part of Fig. 2). From Yan et al. (2021), Remote Sensing.</em></p> </article> </main></body></html>