A Dynamic Cloud Bayes Network-Based Cleaning Method of Multi-Source Unstructured Data

Abstract

Aiming at the problems of data redundancy and data abnormality of multi-source unstructured data such as video, picture, and text in the process of processing quality inspection and equipment status monitoring of discrete intelligent production line, a multi-source unstructured data cleaning method based on dynamic cloud Bayesian network is proposed. We analyze the characteristics of multi-source unstructured data in the processing operation of the discrete intelligent production line and construct a multi-source unstructured data description model. combine dynamic Bayesian network and cloud model theory to design a multi-source unstructured data cleaning framework and processing flow based on dynamic cloud Bayesian network. finally, the feasibility of the proposed method is demonstrated by simulation analysis of arithmetic cases.