yaojingang/geo-citation-lab
GEO experiment data reports and a curated GEO/AEO/AI search paper library.
What it solves
AI搜索(如ChatGPT、Google AI Overview、Perplexity)在选择信源、吸收内容和呈现实体时存在不透明性。这个项目提供公开的实证研究数据和工具,帮助研究者和内容创作者理解AI搜索如何决定哪些内容被引用、如何被吸收,以及Web端和App端在信源呈现上的差异。
How it works
仓库整理了两条研究主线:跨平台实验(覆盖602条受控Prompt,记录引用选择和吸收深度)和中文生成式搜索数据集(CN-GEO,包含超过21万条原始引用记录,来自8个Web/App界面)。所有数据、处理脚本、分析报告和论文PDF被系统化组织,并提供一个轻量Viewer,用户可直接通过AI客户端安装,无需配置复杂环境。此外还包含一个30题在线测试,用于评估对GEO概念的理解。
Who it’s for
面向对AI搜索可见性感兴趣的学者、从事搜索引擎优化(GEO)的从业者、内容创作者,以及希望了解生成式搜索如何选择信源的普通研究者。不需要编程基础即可通过在线报告和Viewer阅读结论,需要深入分析的用户也能获得完整数据与脚本。
Highlights
- 包含214,119条中文引用记录和12个国内AI平台/终端代码,数据规模大且可查询
- 提供跨平台实验的完整管线,涵盖602条Prompt和72维页面特征
- 收录54篇GEO/AEO/AI搜索相关论文PDF,按10个主题分类导航
- 支持一键安装轻量Viewer,或通过AI客户端部署到GitHub Pages
- 提供30题在线能力测试并生成六维诊断报告
- 所有数据使用明确说明限制(如缺少时间戳),保证研究透明性
Related
- Project
ispras/dedocA universal document conversion system that extracts logical structure, tables, and formatting from diverse formats like PDF, Office, and images into a unified output.
- Project
ozekik/awesome-ontologyA curated index of ontology tools, languages, ready-made ontologies (upper/mid/cross/domain), and related resources — not a software project, but a reference list. It groups items by purpose and flags archived or inactive entries.
- Project
facebookincubator/nimbleA columnar file format by Meta designed for large, wide datasets, optimizing storage and access for machine learning training tables and feature engineering.
- Project
jeffhajewski/latticedbLatticeDB is an embedded, single‑file property‑graph database that also provides native HNSW vector similarity search and BM25 full‑text indexing. It offers a unified Cypher‑style query language, ACID transactions, and bindings for Python, TypeScript/Node, Go, and Java, making it suitable for local knowledge‑graph apps, agent memory, and RAG pipelines.