A Machine-Learning-based Study on Data Quality Assessment for an Intelligent Multi-dimensional Information Management System

Abstract

Conventional system data quality assessment methods are based on data characterization and do not take into account the impact of missing and inconsistent data on quality assessment, which results in poor data quality assessment. Therefore, a machine learning-based data quality assessment method for multidimensional information intelligent management system is designed. Comprehensiveness enhancement of multidimensional information intelligent management system data, according to the data of the working conditions of the comprehensive evaluation and supplementation, to ensure the accuracy of the management system data quality assessment. Measure the dimensions of data quality assessment of intelligent management system and formulate reasonable constraint rules for data quality assessment. Based on machine learning, a data quality assessment mechanism is established for the multi-dimensional information intelligent management system, and machine learning models such as random forest, Bayesian algorithm and genetic algorithm are embedded into the multi-dimensional information intelligent management system to realize the automated assessment of data quality. Comparative experiments are used to verify that the assessment results of this method are more accurate and can be applied in real life.