多方法联合预测“四川科技兴村”平台有效数据量  

Forecast valid data of SiChuan-TechRural-Revitalization-Plat based on multi-method integration

作　　者：廖建雄 刘黎立 程小帆 LIAO Jianxiong;LIU Lili;CHENG Xiaofan(Sichuan Boda Nachuan Science and Technology Co.,Ltd.,Chengdu 610041,China;Sichaun Institute of Computer Sciences,Chengdu 610041,China)

机构地区：[1]四川博大纳川科技有限责任公司,四川成都610041 [2]四川省计算机研究院,四川成都610041

出　　处：《技术与市场》2024年第6期41-46,共6页Technology and Market

摘　　要：“四川科技兴村”平台有效数据量预测依赖于现有的移动平均法,而移动平均法在实际应用中无法准确捕捉数据的实时动态变化,导致预测结果未能达到期望的精度和效果。为了更准确地预测“四川科技兴村”平台有效数据量,提出了环同期预测法、次级平台加权平均法和数据留存评估法联合预测的方法。在提出的方法中,环同期预测法在时间序列数据基础上对月度定量预测;次级平台加权平均法完成对月度数据的修正,降低误差;数据留存评估法通过对数据量-有效数据量关系生成每月有效数据预测量;汇总成周期内有效数据预测量。理论分析及验证表明:多方法联合预测提供了“四川科技兴村”平台有效数据量的预测方法。The valid data of SiCHuan-TechRural-Revitalization-Plat relies on the moving average method,which cannot accurately capture the real-time dynamic changes of data in practical applications,resulting in the prediction results not meeting the expected accuracy and effectiveness.In order to effectively forecast valid data of SiCHuan-TechRural-Revitalization-Plat,a multi-method is proposed,which includes a contemporaneous prediction method,secondary platform weighted average method and data retention evaluation method.In the multi-method,the contemporaneous prediction method quantitatively predicts monthly data based on time series data;the secondary platform weighted average method corrects monthly data and reduces errors;the data retention evaluation method generates monthly effective data premeasurements by examining the relationship between data and valid data;summarize into valid data pre measurements within one cycle.Theoretical analysis and experimental results demonstrate that this provides an effective method for forecasting the valid data of SiCHuan-TechRural-Revitalization-Plat.

关 键 词：环同期预测法 次级平台加权平均法 数据留存评估法 移动平均法 多方法联合预测 有效数据量 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论] O213[自动化与计算机技术—计算机科学与技术] F323.3[理学—概率论与数理统计]