| 基于自注意力机制的阅读理解模型 |
| |
| 引用本文： | 张浩宇,张鹏飞,李真真,谭庆平. 基于自注意力机制的阅读理解模型[J]. 中文信息学报, 2018, 32(12): 125-131 |
| |
| 作者姓名： | 张浩宇 张鹏飞 李真真 谭庆平 |
| |
| 作者单位： | 国防科技大学 计算机学院,湖南 长沙 410072 |
| |
| 基金项目： | 国家重点研发计划(2016YFB0200401);新世纪优秀人才,国防科技卓越人才(2017-JCJQ-ZQ-013);湖南省人才计划(2017RS3045) |
| |
| 摘 要： | 机器阅读理解是自然语言处理领域一项得到广泛关注与研究的任务。该文针对中文机器阅读理解数据集DuReader,分析其数据集的特点及难点,设计了一种基于循环神经网络和自注意力机制的抽取式模型Mixed Model。通过设计段落融合等策略,该文提出的模型在DuReader测试集上达到了54.2的Rouge-L得分和49.14的Bleu-4得分。
|
| 关 键 词： | 机器阅读理解 深度学习 自注意力机制 |
Self-attention Based Machine Reading Comprehension |
| |
| ZHANG Haoyu,ZHANG Pengfei,LI Zhenzhen,TAN Qingping. Self-attention Based Machine Reading Comprehension[J]. Journal of Chinese Information Processing, 2018, 32(12): 125-131 |
| |
| Authors: | ZHANG Haoyu ZHANG Pengfei LI Zhenzhen TAN Qingping |
| |
| Affiliation: | School of Computer Science, National University of Defense Technology, Changsha, Hunan 410072, China |
| |
| Abstract: | Machine reading comprehension has attracted concerns in the field of Natural Language Processing. To deal with the Chinese machine reading comprehension data set —DuReader, this paper presents an extractive language model called Mixed Model with multiple strategies including recurrent neural network, paragraph fusion and self-attention mechanism. The proposed method achieves a Rouge-L score of 54.2 and a Bleu-4 score of 49.14 on the DuReader test set. |
| |
| Keywords: | machine reading comprehension deep learning self attention mechanism |
|
| 点击此处可从《中文信息学报》浏览原始摘要信息 |
|
点击此处可从《中文信息学报》下载全文 |
|
