📄 中文摘要
英国规划部门面临《规划法》和《数据保护法》之间的立法冲突,前者要求公众获取申请文件,而后者则要求保护个人信息。这种情况导致在处理大量文件时工作量繁重,使规划官员不得不将精力转向行政任务,增加了法律合规风险。为了解决这些挑战,提出了一种集成的人工智能系统,该系统能够自动识别和编辑个人信息,从规划文件中提取关键元数据,并分析建筑图纸中的特定特征。该系统采用AI-in-the-Loop(AI2L)设计,所有建议均需进行审核。
📄 English Summary
Automating Document Intelligence in Statutory City Planning
UK planning authorities are confronted with a legislative conflict between the Planning Act, which mandates public access to application documents, and the Data Protection Act, which necessitates the safeguarding of personal information. This situation results in a labor-intensive workload for processing large volumes of documents, diverting planning officers to administrative tasks and increasing legal compliance risks. An integrated AI system is proposed to tackle these challenges. The system automates the identification and redaction of personal information, extracts key metadata from planning documents, and analyzes architectural drawings for specified features. It operates with an AI-in-the-Loop (AI2L) design, requiring all suggestions to be reviewed.
Powered by Cloudflare Workers + Payload CMS + Claude 3.5
数据源: OpenAI, Google AI, DeepMind, AWS ML Blog, HuggingFace 等