基于改进GRU与MVC设计模式的数据智能分析算法  

Intelligent data analysis algorithm based on improved GRU and MVC design patterns

作　　者：牛洁 NIU Jie(Xi’an Aeronautical Polytechnic Institute,Xi’an 710089,China)

机构地区：[1]西安航空职业技术学院,陕西西安710089

出　　处：《电子设计工程》2024年第10期25-29,共5页Electronic Design Engineering

基　　金：全国教育科学“十三五”规划2020年度教育部重点课题(DJA200310);西安航空职业技术学院2022年度科研计划项目(21XHZX-01)。

摘　　要：针对传统财务异常数据检测方法效率较低、准确度差且坏账率高的问题,文中基于改进的人工智能算法提出了一种异常数据检测方法。由于高维异常数据难以分析,先用孤立森林算法将其剔除,再将处理后的数据经过双向GRU算法的训练,挖掘出数据的时序性特征。对于训练后数据分类准确度较低的问题,通过注意力机制对数据特征权重进行排序,从而得到最终的分类结果。基于MVC设计了软件架构进行实验测试,该算法的训练总时长明显低于对比算法,RMSE及MAPE指标相较Bi-LSTM算法低0.2%和0.15%,且准确率、召回率与F1值在对比算法中也为最优。Aiming at the problems of low efficiency,poor accuracy and high bad debt rate of traditional financial abnormal data detection methods,this paper proposes an abnormal data detection algorithm based on improved artificial intelligence algorithm.Because high⁃dimensional abnormal data is difficult to analyze,the isolated forest algorithm is used to remove it first,and then the processed data is trained by Bi-GRU algorithm to mine the temporal characteristics of the data.For the low accuracy of data classification after training,the data feature weights are sorted through the attention mechanism to obtain the final classification results.The software architecture is designed based on MVC for experimental testing.The total training time of this algorithm is significantly lower than that of the comparison algorithm.The RMSE and MAPE indicators are 0.2%and 0.15%lower than that of Bi-LSTM algorithm,and the accuracy,recall and F1 value are also optimal in the comparison algorithm.

关 键 词：异常数据分析 孤立森林算法 双向GRU 注意力机制 MVC设计 大数据 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]