A Prediction Model Combining Convolutional Neural Network and LSTM Neural Network
Ao Liu, Jing Li, Han Ye · 2023
Existing models such as Recurrent Neural Network (RNN) and Long Short Term Memory (LSTM) can make immediate predictions, while making long-term predictions is more difficult. In order to make better prediction, this paper proposes a hybrid model consisting of convolutional model and Long Short Term Memory model (CNN-LSTM) to make prediction.CNN-LSTM hybrid neural network firstly extracts the spatial structure features of the data by CNN. Secondly, the time series features in the data are extracted by using LSTM, and finally the prediction is effective in a short time. By comparing the experimental prediction results and evaluation indexes, it is verified that the CNN-LSTM hybrid network model has more accurate prediction accuracy and stability than some traditional prediction models, which has high reference value.