Student score prediction method of convolutional long and short term network based on attention mechanism

The invention relates to a student score prediction method of a convolutional long and short term network based on an attention mechanism, which predicts a professional course score by utilizing a pre-repair course score of a student and can be applied to early warning before the professional course starts. The method comprises the following steps: inputting obtained pre-repair course score data into a convolutional long-short-term memory network layer for feature learning; the learned sequence features are input to an attention mechanism layer, and different weights are distributed to different pre-repair course score features; the learned sequence features and weights generated by the attention mechanism are sent to a merging layer for feature merging; the merged features are sent to two full-connection network layers for classification, and the final output is a prediction result; the long-short-term memory network is a prediction network capable of processing time sequence data, the ...