Finance & Economics AI
收藏资源简介:
This dataset is a **substantial, well-structured historical financial corpus** comprising over **2 million records** drawn from 10 authoritative financial publications spanning from **1843 to the early 1930s**. It has undergone rigorous OCR processing and paragraph reconstruction, resulting in excellent core data completeness. ****Please refer to documentation for AI readiness training report.*** Ten publications. Tens of thousands of pages. 75 years of financial history (1855‑1930). Cleaned. Structured. Bias‑audited. <p><br/></p> **Included publications:** <p><br/></p> 🏦 Lloyds Bank Monthly (1930) – British banking perspective, corporate finance, international trade 📊 The Analyst (1913-1930) – Financial analysis, market commentary, investment research 💳 Business Credit (1893‑1930) – Commercial credit, trade finance, corporate lending 📈 Trusts & Estates (1904‑1930) – Trust administration, estate planning, fiduciary law 🏛️ NY Reserve Monthly Review (1919‑1930) – Federal Reserve Bank of New York economic commentary 🗽 Citibank Monthly Letter (1914‑1930) – Institutional banking analysis, economic conditions 🌍 The Economist (1843‑1930) – Global economics, finance, politics, trade (flagship) 📜 Commercial & Financial Chronicle (1865‑1930) – America's first national business weekly 📚 Credit World (1914‑1930) – Consumer credit, credit reporting, retail finance 🔒 Commercial & Financial Chronicle (1865‑1930) – Trust company operations, corporate trusteeship <p><br/></p> **Use this archive to:** <p><br/></p> - Train financial AI models on a century of banking, credit, and trust literature - Extract financial entities: interest rates, bond yields, credit terms, trust structures - Study the evolution of central banking, commercial credit, and international finance - Support research in financial history, economic policy, and banking regulation <p><br/></p> All content is professionally OCR‑cleaned, structured as JSONL, and bias‑audited with historical context notices. 100% public domain (pre‑1930). Snowflake‑optimized. <p><br/></p> **Sample Cortex Agent Prompts:** 1. *What are the articles about banking or credit in the finance archive?* <p><br/></p> 2. *How many articles does each publication have?* <p><br/></p> 3. *What are the titles of finance articles that discuss interest rates or discount rates?* <p><br/></p> Row count: 2,016,509 Time period: 1843‑1930 Publications: 11 <p><br/></p>
数据集概述:Finance & Economics AI
该数据集是一个大规模、结构化的历史财务语料库,由Devin Media Corp.提供,AI训练准备度评分为84/100,被认为是适合RAG和金融AI应用的强候选数据集。
核心数据特征
- 记录数量:超过200万条记录(2,016,509行)
- 时间跨度:1843年至1930年代初
- 数据来源:来自10种权威金融出版物,经过专业OCR处理和段落重建,数据完整度优秀。
- 数据格式:经过专业OCR清洗,结构化为JSONL格式,已进行偏差审计,并附带历史背景说明。
- 数据状态:100%公共领域(1930年前内容)。
- 数据交付:通过安全共享方式交付,每月刷新。
包含的出版物
| 出版物 | 时间范围 | 内容侧重 |
|---|---|---|
| 🏦 Lloyds Bank Monthly | 1930 | 英国银行业视角,企业金融,国际贸易 |
| 📊 The Analyst | 1913-1930 | 金融分析,市场评论,投资研究 |
| 💳 Business Credit | 1893-1930 | 商业信贷,贸易融资,企业借贷 |
| 📈 Trusts & Estates | 1904-1930 | 信托管理,遗产规划,信托法律 |
| 🏛️ NY Reserve Monthly Review | 1919-1930 | 纽约联邦储备银行经济评论 |
| 🗽 Citibank Monthly Letter | 1914-1930 | 机构银行分析,经济状况 |
| 🌍 The Economist | 1843-1930 | 全球经济,金融,政治,贸易(旗舰刊物) |
| 📜 Commercial & Financial Chronicle | 1865-1930 | 美国第一本全国性商业周刊 |
| 📚 Credit World | 1914-1930 | 消费者信贷,信用报告,零售金融 |
| 🔒 Commercial & Financial Chronicle | 1865-1930 | 信托公司运营,公司信托管理 |
注:出版物总计为11个。
数据结构与数据字典
数据集包含一个名为 FINANCE_ARCHIVE 的表,其关键字段如下:
- ISSUE (Varchar): 出版物期号
- TITLE (Varchar): 文章标题
- AUTHOR (Varchar): 作者
- TYPE (Varchar): 内容类型(例如:article)
- TEXT (Varchar): 全文文本
- INGESTION_DATE (Timestamp_NTZ): 数据摄取日期
业务应用场景
- 模型开发 (Model Development):在银行、商业信贷、信托管理和央行政策文献上训练特定领域的LLM,支持模型预训练和微调,用于金融分析、信用风险评估和经济预测。
- 实体识别 (Entity Recognition):从19-20世纪的金融文本中提取利率、债券收益率、信用条款、信托结构、银行名称、央行政策和经济指标等金融实体。
- 文本摘要 (Text Summarization):利用全文金融文章与原始标题形成的自然源-摘要对,训练模型生成冗长市场评论、信用报告和经济分析的简明摘要。
定价模式
- 该数据集(Finance AI Historical Banking and Credit Archive)的定价信息未在页面详细列出,需联系获取。
用例示例(SQL查询)
- 查看数据集内容:选择
ISSUE, TITLE, AUTHOR,筛选类型为article,限制10条。 - 按主题查找文章:使用
ILIKE搜索文本中包含banking或credit的内容。 - 按出版物统计文章数量:使用
SPLIT_PART函数从字段中提取出版物名称进行分组统计。 - 搜索利率内容:使用
ILIKE搜索文本中包含interest rate或discount rate的内容。



