基于DFG-GEPRIS权威数据的德英双语德国科学基金会资助项目数据集

本产品基于SQLite/Excel存储，将德国DFG数据重构为由“节点表”与“边表”构成的关系型知识图谱。
1. 组织方式与字段规范：
-项目表 (Projects)：主键为Project_ID。存有标题、摘要 (String)、起止年份 (Int)、资助类型、学科、项目介绍及Is_China_Related (Bool)等属性 。
-人员表 (Persons)：主键为Person_ID。含姓名、地址及溯源 URL (String) 。
-机构表 (Institutions)：主键为Institution_ID。含机构名、地址及溯源 URL。
-成果表 (Publications)：记录论文标题及发表年份 (Int)等 。
-边表 (Edges)：含“项目-人员”与“项目-机构”表 。通过主外键关联项目ID与人员/机构 ID，定义角色关系 。
2. 计算存储逻辑：
采用异步高并发引擎抓取，利用SQLite WAL模式确保数据写入的原子性 。通过正则表达式及降维算法将非结构化数据扁平化为二阶关系表，支持社会网络分析(SNA)与面板数据运算 。

This product is developed based on SQLite/Excel storage, and reconstructs the German DFG data into a relational knowledge graph consisting of "node tables" and "edge tables".

1. Organization and Field Specifications:
- Projects Table: The primary key is Project_ID. It stores attributes such as title, abstract (String), start and end years (Int), funding type, discipline, project introduction, and Is_China_Related (Bool).
- Persons Table: The primary key is Person_ID. It includes name, address, and source URL (String).
- Institutions Table: The primary key is Institution_ID. It includes institution name, address, and source URL.
- Publications Table: Records paper titles, publication years (Int), and other relevant information.
- Edges Table: Includes "Project-Person" and "Project-Institution" tables. It associates project IDs with person/institution IDs via primary and foreign keys to define role relationships.

2. Computational and Storage Logic:
An asynchronous high-concurrency engine is adopted for data crawling. The SQLite WAL mode is utilized to ensure the atomicity of data writing. Unstructured data is flattened into second-order relational tables through regular expressions and dimensionality reduction algorithms, supporting social network analysis (SNA) and panel data operations.

搜集汇总

背景与挑战

背景概述

该数据集基于DFG-GEPRIS权威数据源，收录了德国科学基金会（DFG）资助项目的相关信息，并以德英双语形式呈现。它主要面向科研管理、项目分析或资助研究等领域，提供了结构化的项目数据支持。