Network flow prediction method based on bidirectional long-short-term memory recurrent neural network

The invention discloses a network flow prediction method based on a bidirectional long-short-term memory recurrent neural network, and belongs to the field of computer networks. The method includes processing the collected network traffic original data by using a sliding window technology, and then performing normalization processing; initializing parameters of the bidirectional long-term and short-term memory recurrent neural network prediction model; inputting the normalized network flow data into the initialized bidirectional long-term and short-term memory recurrent neural network model; training a bidirectional long-term and short-term memory recurrent neural network to perform bidirectional learning; the overall characteristics of the network flow are mined and memorized; judging whether the training count value reaches the training times or not; training whether a target reaches a set error requirement or not; according to the method, bidirectional learning is carried out on thenetwork flow sequence, a