司法涉诉风险大数据

司法涉诉风险大数据集，基于司法涉诉、司法执行、司法失信等数据，通过数据挖掘、机器学习、统计分析、成分分析、聚类分析、决策树等大数据技术工具，提炼出司法诉讼的风险因素、风险损失、风险要素等，以及发现数据之间的关联性和规律性。主要应用场景为：金融机构风险管理、政府部门风险管理、企业部门风险管理等。风险分数分值越大，风险越大。风险分数计算公式：风险分数 = (w1 * z_score(Exec_Count) + w2 * z_score(Exec_Amount) + w3 * z_score(Pend_Count) + w4 * z_score(Pend_Amount)) / sum(w)
Exec_Count表示被执行总次数，Exec_Amount表示被执行总金额，Pend_Count表示未结案数量，Pend_Amount表示未结案的金额，z_score()函数表示标准化函数，将每个特征值转化为标准正态分布下的z值。w1、w2、w3、w4分别表示每个特征的权重，权重根据业务经验和历史数据进行调整，sum(w)表示权重之和。

Judicial Litigation-related Risk Big Dataset is developed based on data from judicial litigation-related matters, judicial enforcement, judicial dishonesty and other sources. It leverages big data technical tools including data mining, machine learning, statistical analysis, component analysis, cluster analysis, decision trees and others to extract risk factors, risk losses and risk elements of judicial litigation, and to identify the correlations and regularities underlying the data. Its core application scenarios cover risk management for financial institutions, government departments and enterprise departments. The larger the risk score value, the higher the corresponding risk. The risk score is calculated via the following formula:
Risk Score = (w1 * z_score(Exec_Count) + w2 * z_score(Exec_Amount) + w3 * z_score(Pend_Count) + w4 * z_score(Pend_Amount)) / sum(w)
Where Exec_Count denotes the total number of enforcement actions, Exec_Amount denotes the total enforcement amount, Pend_Count denotes the number of pending cases, and Pend_Amount denotes the amount of pending cases. The z_score() function is a standardization function that transforms each feature value into a z-score under the standard normal distribution. w1, w2, w3 and w4 represent the weights of each respective feature, which are adjusted based on business experience and historical data, while sum(w) refers to the sum of all the weights.

搜集汇总

数据集介绍

特点

司法涉诉风险大数据集包含近100万条数据，每日更新，涵盖司法涉诉、执行和失信等数据，通过大数据技术提炼风险因素。主要应用于金融机构、政府部门和企业的风险管理，风险分数通过加权标准化特征计算得出，分值越大风险越高。