SAP/SALT-KG

该数据集来自论文《SALT-KG: 企业表格语义感知学习基准测试》，发表于EurIPS25表格表示研讨会。SALT-KG在SALT关系预测基准测试的基础上，通过将其多表事务数据与一个结构化的操作业务知识（以元数据知识图谱形式表示）连接起来，构建了一个用于企业表格语义感知学习的基准测试。该元数据知识图谱捕获了字段级描述、关系依赖和业务对象类型。数据集包含：4个来自SALT基准测试的关系表（带有事务数据）、一个包含字段级描述、关系依赖和业务对象类型的元数据知识图谱（OBKG）、规范元数据JSON文件（data/salt-kg/salt-kg.json）、为Hugging Face查看器生成的辅助文件（data/salt-kg/salt-kg-summary.parquet和data/salt-kg/salt-kg-fields.parquet），以及为基准测试表提供的训练和测试分割（如果提供），以及用于直接浏览的完整表parquet文件。This repository contains the dataset from our paper SALT-KG: A Benchmark for Semantics-Aware Learning on Enterprise Tables, presented at EurIPS25 Table Representation Workshop. Building upon the SALT benchmark for relational prediction, SALT-KG extends SALT by linking its multi-table transactional data with a structured Operational Business Knowledge represented in a Metadata Knowledge Graph (OBKG) that captures field-level descriptions, relational dependencies, and business object-types. The dataset consists of 4 relational tables with transactional data, a Metadata Knowledge Graph (OBKG) with field-level descriptions, relational dependencies, and business object types, a canonical metadata JSON file at `data/salt-kg/salt-kg.json`, generated Hugging Face viewer helper files at `data/salt-kg/salt-kg-summary.parquet` and `data/salt-kg/salt-kg-fields.parquet`, and train and test splits for the benchmark tables where provided, plus full-table parquet files for direct browsing.