Couplet generation model, couplet generation method and device, computer equipment and medium

Abstract

The invention belongs to the field of artificial intelligence, and particularly relates to a couplet generation model, a couplet generation method and device, computer equipment and a medium. The method comprises the steps of obtaining training data containing couplet data and coded information associated with the couplet data; performing word segmentation on the training data based on a preset rhythm structure to generate a plurality of keywords with word structures; generating system keywords in one-to-one correspondence with the keywords, and combining the keywords and the system keywords based on keyword combination characteristics to generate multiple keyword combinations; and converting each keyword combination into a group of combined sequence samples, inputting all the combined sequence samples into a seq2seq model, and training the seq2seq model according to the coded information to obtain a couplet generation model based on the seq2seq model. The invention also relates to a block chain technology, and the keyword combinations are stored in a block chain. According to the invention, the antithetical property of antithetical couplets and the satisfaction of users can be improved.

Description

technical field [0001] The invention relates to the field of intelligent decision-making of artificial intelligence, in particular to a couplet generation model, a couplet generation method, device, computer equipment and media. Background technique [0002] The couplets are one of the traditional Chinese cultures. The couplets have the characteristics of neat antithesis and harmonious balance. They are a unique art form of the Chinese language with one word and one sound, and they are treasures of Chinese traditional culture. Couplets are also a common means for people to express blessings in scenes such as traditional festivals and opening ceremonies. It can be seen that couplets are widely used. But creating a couplet with neat antithesis and expressing appropriate text emotion is not an easy task for ordinary people without literary background. [0003] At present, the traditional couplet data processing method requires the user to input the upper couplet in the machine...

AI Technical Summary

Problems solved by technology

But creating a couplet with neat antithesis and expressing appropriate text emotion is not an easy task for ordinary people without literary background

[0003] At present, the traditional couplet data processing method requires the user to input the upper couplet in the machine, and then let the machine generate a neat lower couplet with the upper couplet. Traditionally, this couplet data processing method reduces the difficulty for users to create the entire couplet, but for ordinary people without literary skills, it is still difficult to create the first couplet of the couplet

At the same time, couplets need to pay attention to antithesis and text emotion, but the literary level of each user is not consistent. If the words and sentences in the first couplet input by the user do not use appropriate antithetical words, the second couplet generated by the machine will not conform to the grammatical rules, that is, all The generated couplets will also fail to reflect antithesis and text emotion, so it is difficult to create couplets that satisfy users with the current traditional couplet data processing methods

Images