Improving data quality using a deep learning network
Chulhyun Hwang, Kyouhwan Lee, Hoe-Kyung Jung · Indonesian Journal of Electrical Engineering and Computer Science · 2020
IoT data is collected in real time and is treated as highly reliable data because of its high precision. However, it often exhibits incomplete values for reasons such as sensor aging and failure, poor operating environment, and communication problems. The characteristics of IoT data transmitted with high precision and time series are suitable to use LSTM, which is one kind of RNN. In this paper, when applying LSTM to data quality improvement in IoT environment where data are collected simultaneously from several sensors, it is suggested that it is effective to construct LSTM individually for each sensor accuracy.