| 融合深度学习和链路预测的细粒度技术预测研究——以合成生物技术为例 |
| |
| 引用本文： | 胡雅敏,吴晓燕,廖兴滨,钱杨舸,陈方.融合深度学习和链路预测的细粒度技术预测研究——以合成生物技术为例[J].图书情报工作,2022,66(24):92-103. |
| |
| 作者姓名： | 胡雅敏 吴晓燕 廖兴滨 钱杨舸 陈方 |
| |
| 作者单位： | 1. 中国科学院成都文献情报中心 成都 610299;2. 中国科学院大学经济与管理学院图书情报与档案管理系 北京 100190;3. 中国科学院成都计算机应用研究所 成都 610299 |
| |
| 基金项目： | 本文系中国科学院成都文献情报中心2021年创新基金青年项目"基于知识基因的领域创新路径分析框架研究"(项目编号:E1Z0000202)研究成果之一。 |
| |
| 摘 要： | 目的/意义]面向专利文本进行更细粒度的技术实体识别和技术预测,利于更详细地把握专利技术布局与趋势。方法/过程]首先利用深度学习方法自动识别专利技术术语类实体,通过实验对比多组深度学习算法的优劣。其次,提出新的半监督标注和自定义标注方案,提高人工标注效率。最后,执行训练得到的最优模型,结合链路预测方法,对合成生物技术进行细粒度的技术预测。结果/结论]实证结果表明RoBERTa-BiLSTM-CRF模型更适用于语义复杂的专利技术实体识别,F1值可达到86.8%,技术识别结果比传统IPC分析方法更精细。同时,细粒度的技术预测结果表明,合成生物学的合成方法在不断改进创新,合成物研究向合成燃料发展。
|
| 关 键 词： | 技术术语识别 深度学习 技术预测 合成生物 |
| 收稿时间： | 2022-05-18 |
| 修稿时间： | 2022-09-26 |
Research on Fine-Grained Technology Prediction Based on Deep Learning and Link Prediction: Take Synthetic Biology as an Example |
| |
| Hu Yamin,Wu Xiaoyan,Liao Xingbin,Qian Yangge,Chen Fang.Research on Fine-Grained Technology Prediction Based on Deep Learning and Link Prediction: Take Synthetic Biology as an Example[J].Library and Information Service,2022,66(24):92-103. |
| |
| Authors: | Hu Yamin Wu Xiaoyan Liao Xingbin Qian Yangge Chen Fang |
| |
| Institution: | 1. Chengdu Library and Information Center, Chinese Academy of Sciences, Chengdu 610299;2. Department of Library, Information and Archives Management, School of Economics and Management, University of Chinese Academy of Sciences, Beijing 100190;3. Chengdu Information Technology of Chinese Academy of Sciences CO., LTD, Chengdu, 610299 |
| |
| Abstract: | Purpose/Significance] It is beneficial to grasp the layout and trend of patent technology by identifying technical entities and predicting technology with finer granularity for patent texts. Method/Process] The deep learning method was used to automatically identify patent technology terms entities, and the advantages and disadvantages of several groups of deep learning algorithms were compared by empirical analysis. At the same time, new semi-supervised labeling and self-defined labeling schemes were proposed to improve the efficiency of manual labeling. Finally, the optimal model obtained by training was implemented, and the fine-grained technical prediction of synthetic biotechnology was made by combining the link prediction method. Result/Conclusion] The empirical results show that RoBERTa-BiLSTM-CRF model is more suitable for the recognition of patent technical terms with complex semantics, and the F1 value reaches 86.8%. The technical recognition result is more detailed than the traditional IPC analysis method. The fine-grained technical prediction shows that the synthetic methods of synthetic biology are constantly improving and innovating, and the synthetic research is developing towards synthetic fuels. |
| |
| Keywords: | technology terms recognition deep learning technology prediction synthetic biology |
|
| 点击此处可从《图书情报工作》浏览原始摘要信息 |
| 点击此处可从《图书情报工作》下载免费的PDF全文 |
|
