莞数金融×企业公积金衍生数据集

数据集基于东莞市住房公积金原始缴存明细数据，严格按照《广东省公共数据管理办法》及金融监管数据安全规范进行脱敏清洗后，利用Python及Spark大数据处理框架，构建多维度时间窗口统计与趋势分析模型，生成了一系列具有风险识别价值的高度浓缩型智力字段，包括稳定性评估类指标、合规性判断类指标、成长性分析类指标。该数据集可有效应用于金融机构信贷风控场景（如企业贷款资质评估、小微企业授信额度测算）、政府监管部门风险预警（如企业经营异常监测、区域性金融风险识别）及人力资源服务机构的信用评价等场景，显著提升了原始公积金数据在金融风控领域的应用价值与决策效率。

This dataset is developed based on the original contribution detail data of the housing provident fund in Dongguan City. After strict desensitization and cleaning in accordance with the Guangdong Provincial Public Data Management Measures and financial regulatory data security specifications, we utilized Python and Spark big data processing frameworks to construct a multi-dimensional time-window statistics and trend analysis model, generating a series of highly condensed intelligent-derived fields with risk identification value, including stability assessment indicators, compliance judgment indicators, and growth analysis indicators. This dataset can be effectively applied to credit risk control scenarios of financial institutions, such as corporate loan qualification assessment and credit limit calculation for small and micro enterprises, risk early warning for government regulatory departments including abnormal enterprise operation monitoring and regional financial risk identification, as well as credit evaluation scenarios for human resource service institutions, which significantly enhances the application value and decision-making efficiency of the original housing provident fund data in the financial risk control field.

搜集汇总

背景与挑战

背景概述

该数据集基于东莞市住房公积金原始缴存明细数据，通过Python和Spark框架构建多维度时间窗口统计与趋势分析模型，生成稳定性评估、合规性判断和成长性分析三类衍生指标。数据集经过严格脱敏和合规处理，适用于金融机构信贷风控、政府监管风险预警及人力资源信用评价等场景，显著提升了公积金数据在金融风控领域的应用价值。