基于序列到序列神经网络模型的古诗自动生成方法

Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-10-11 Cooperative journals: 《计算机应用研究》

黄文明 卫万成 邓珍荣

Abstract： Computer poetry generation is our first step towards computer writing. At present, there are many problems in computer poetry writing, such as unclear theme, the content of poetry is inconsistent with the writing intention. For improving these problems, this paper follows the process of writing poem by the ancient Chinese poet and proposes a method for generating Chinese poetry with two stages. The first stage extracts the outline. During this process, this paper use TextRank algorithm to extract keywords from user input text, and propose an attention-based sequence to sequence neural network model for expanding keyword. The second stage generates each line of poem based on the outline of poem. During this process, this paper proposed a sequence to sequence neural network model with dual-encoding and attention mechanism for generating poem. At the end, this paper verified the effectiveness of our approach by evaluation. Compared with baseline approach, the theme of the poem generated by our approach is more explicit, and the contents expressed by the poem are more consistent with the writing intention.

YES

 Hits 2234  Downloads 1249  Comment 0