宁海县住宿餐饮业法人用水用户信用评估数据

为贯彻落实信用信息基础库的建设，采集在经济活动中产生的用水数据和缴费行为数据，通过数据挖掘技术，评估用户的信用水平。通过信用评估，相关部门能够对高风险企业加强巡查和监管，推动企业进行绿色生产;促进水务、环保、税务等部门之间的数据共享与协同，形成统一的监管机制，提高管理效率。1.收集用户历史缴费数据，包括上一年度用水总量、上一年度费用、上一月份用水总量、上一年度同比用水总量、上一月份费用产生日期、开始缴费日期、逾期日期、实际缴费日期 2.根据数据分析用户行为失信特征数据（逾期次数A1、最大逾期天数A2、平均逾期天数A3）、按期缴费行为守信特征数据（按时缴费比例B1）、用水行为特征数据（上一年度平均每月用水总量、同比差异百分比C1=(上一月份用水总量-上一年度同比用水总量/12)/(上一年度同比用水总量/12)、平均每月用水差异百分比C2=(上一月份用水总量-上一年度平均每月用水总量)/上一年度平均每月用水总量） 3.将上述已归纳的用户失信特征、守信特征、用气特征进行计算,分值P=(100-A1-A2-A3)*B1+C1+C2，将用户信用等级划分为优秀(P≥100)、良好(80≤P＜100)、中等(50≤P＜80)、差(P<50)。To advance the construction of the basic credit information database, water usage data and payment behavior data generated during economic activities are collected, and users' credit levels are evaluated via data mining technology. Through credit assessment, relevant departments can strengthen inspections and supervision over high-risk enterprises, promote enterprises to adopt green production practices, facilitate data sharing and collaboration across water affairs, environmental protection, tax and other departments, establish a unified supervision mechanism, and improve management efficiency. 1. Collect users' historical payment data, including total water consumption in the previous year, total expenses in the previous year, total water consumption in the last month, year-on-year total water consumption in the previous year, expense generation date of the last month, start payment date, overdue date, and actual payment date. 2. Analyze users' credit default behavior characteristic data (number of overdue payments A1, maximum overdue days A2, average overdue days A3), credit-worthy behavior characteristic data for timely payment (timely payment ratio B1), and water usage behavior characteristic data (average monthly water consumption in the previous year, year-on-year difference percentage C1=(total water consumption in last month - year-on-year total water consumption in the previous year / 12) / (year-on-year total water consumption in the previous year / 12), average monthly water consumption difference percentage C2=(total water consumption in last month - average monthly water consumption in the previous year) / average monthly water consumption in the previous year) through data analysis. 3. Calculate using the summarized user credit default characteristics, credit-worthy characteristics, and water usage characteristics: the credit score P = (100 - A1 - A2 - A3) * B1 + C1 + C2. Divide user credit levels into four categories: Excellent (P≥100), Good (80≤P<100), Medium (50≤P<80), and Poor (P<50).