Construction method of hybrid neural network model for dialogue generation

Abstract

The invention discloses a construction method of hybrid neural network model for dialogue generation. The construction method of hybrid neural network model for dialogue generation includes the steps: acquiring a data set in a mode of dialogue statement pairs, and constructing a glossary; generating a word embedded table; initializing the convolution neural network with special structure, generating a vocabulary recommending table corresponding to the input statement, determining whether real output is provided, and if so, training the parameters of the convolution neural network in the step; initializing the recurrent neural network with special structure, using the last step to output, generating a vocabulary identity list with word order, determining whether real output is provided, and if so, training the parameters of the recurrent neural network in the step; after the training result satisfies the set index, saving the glossary and the word embedded table, and saving the parameters of the convolution neural network and the recurrent neural network, thus completing construction of the whole model. The construction method of hybrid neural network model for dialogue generation solves the problems that a current neural network dialogue model is slow in the training speed, low in the accuracy and general in statement generation because the glossary is too long.

Description

technical field [0001] The invention relates to the fields of artificial intelligence, neural network and natural language processing, in particular to a method for constructing a hybrid neural network model for dialogue generation. Background technique [0002] In the field of natural language processing, dialogue generation has always been a subject of widespread concern and difficulty. In the field of artificial intelligence, whether a machine can use natural language to communicate like a human is one of the important criteria for judging whether artificial intelligence is powerful. Before neural networks are widely used, the main methods of dialogue generation tasks rely on statistical methods and retrieval methods, which are not only limited to specific fields, but also require artificially set rules for guidance in most tasks. After the emergence of neural networks, especially after the wide application of convolutional neural networks in the field of image processin...

AI Technical Summary

Problems solved by technology

In the dialogue task, the neural network model needs to face a huge number of categories, and the category distribution is uneven, accompanied by rare classification but extremely valuable phenomenon, it is difficult for the neural network model to predict such vocabulary

The current neural network model does not solve the above problems. Although the existing related models reduce the number of classifications during training through random sampling, this method leads to unsatisfactory prediction results, and the prediction effect depends on the selection of sampling distribution.

In addition, the existing neural network model is difficult to divide into modules and needs to be trained and used as a whole. If you need to change a part of it to achieve some special purpose, the modification is troublesome and requires retraining, resulting in a waste of existing results

Images