A Novel Big Data Cleaning Algorithm Based On Edge Computing In Industrial Internet of Things
Mao Chen, Wen Chen, Yongqi Zhu · 2021
The Industrial Internet of Things(IIoT) is a revolution that is changing the face of industry. It brings opportunities and also challenges. Due to the harsh sensor environment in the industry, the collected big data is not credible, which seriously affects the judgment and feedback of the cloud. Traditional data cleaning relying on sensor nodes is not enough to process big data, while mobile edge computing can provide a good solution. The paper proposes a data cleaning solution based on mobile edge nodes. First,we obtain the training data of the cleaning model. Second, we use the isolation forest (iForest) anomaly detection method at the edge nodes. Experimental results show that the scheme improves the efficiency of data cleaning and also ensures the reliability and integrity of the data.