|
|||||
|
|
| 基于藏文Albert预训练语言模型的图采样与聚合实体关系抽取 | |
| 引用本文： | 于韬,尼玛次仁,拥措,尼玛扎西.基于藏文Albert预训练语言模型的图采样与聚合实体关系抽取[J].中文信息学报,2022,36(10):63-72. |
| 作者姓名： | 于韬 尼玛次仁 拥措 尼玛扎西 |
| 作者单位： | 1.西藏大学 信息科学技术学院,西藏 拉萨 850000; 2.西藏大学 西藏自治区藏文信息技术人工智能重点实验室,西藏 拉萨 850000; 3.西藏大学 藏文信息技术教育部工程研究中心,西藏 拉萨 850000 |
| 基金项目： | 科技部重点研发计划重点专项(2017YFB1402200);西藏自治区科技创新基地自主研究项目(XZ2021JR002G) |
| 摘 要： | 实体关系抽取任务是对句子中实体对间的语义关系进行识别。该文提出了一种基于Albert预训练语言模型结合图采样与聚合算法(Graph Sampling and Aggregation, GraphSAGE)的实体关系抽取方法,并在藏文实体关系抽取数据集上实验。该文针对藏文句子特征表示匮乏、传统藏文实体关系抽取模型准确率不高等问题,提出以下方案: ①使用预先训练的藏文Albert模型获得高质量的藏文句子动态词向量特征; ②使用提出的图结构数据构建与表示方法生成GraphSAGE模型的输入数据,并通过实验证明了该方法的有效性; ③借鉴GraphSAGE模型的优势,利用其图采样与聚合操作进行关系抽取。实验结果表明,该文方法有效提高了藏文实体关系抽取模型的准确率,且优于基线实验效果。 |
| 关 键 词： | 藏文 实体关系抽取 Albert GraphSAGE |
| 收稿时间： | 2022-02-05 |
Graph Sampling and Aggregation Entity Relation Extraction Based on Tibetan Albert Pre-trained Language Model |
|
| YU Tao,Nyima Ciren,YONG Cuo,Nyima Trashi.Graph Sampling and Aggregation Entity Relation Extraction Based on Tibetan Albert Pre-trained Language Model[J].Journal of Chinese Information Processing,2022,36(10):63-72. | |
| Authors: | YU Tao Nyima Ciren YONG Cuo Nyima Trashi |
| Affiliation: | 1.College of Information Science and Technology, Tibet University, Lhasa, Tibet 850000, China;2.Key Laboratory of Tibetan Information Technology and Artificial Intelligence of Tibet, Tibet University, Lhasa, Tibet 850000, China;3.Engineering Research Center of Tibetan Information Technology, Ministry of Education, Tibet University, Lhasa, Tibet 850000, China |
| Abstract: | Entity relation extraction task is to recognize relations between different entities in sentences. This paper proposes a Graph Sampling and Aggregation (GraphSAGE) Tibetan entity relation extraction method based on Albert pre-trained model. The language model is used to obtain high quality sentence features. Input data for the GraphSAGE model is generated by the graph structure data construction and representation method designed in this paper. The experimental results show that our method is effective and superior to the baseline methods. |
| Keywords: | Tibetan entity relation extraction Albert GraphSAGE |
| 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 | |
