| 基于深度学习的虚拟学术社区智能问答研究 |
| |
| 引用本文： | 徐彤阳,滕琦.基于深度学习的虚拟学术社区智能问答研究[J].情报杂志,2021(4):163-169. |
| |
| 作者姓名： | 徐彤阳 滕琦 |
| |
| 作者单位： | 山西财经大学信息学院 |
| |
| 基金项目： | 2020年度山西省回国留学人员科研资助项目“基于数字人文的山西省城市记忆资源开发策略研究”(编号:2020-098);2019年度山西省研究生教育改革研究课题“元素养教育下融入学术情景的研究生信息检索课程教学模式探索”(编号:2019JG140);2019年山西财经大学教改课题“新媒体时代基于元素养的信息检索课程教学模式探讨”(编号:2019218)的研究成果之一。 |
| |
| 摘 要： | 目的/意义]作为科研人员知识获取、交流与利用的重要场所,虚拟学术社区资源具有数量多、价值高等特征,探索智能问答服务在社区的理论、方法与应用,对推动学术资源高效检索与利用,促进知识服务创新具有重要意义。方法/过程]梳理虚拟学术社区研究现状与目前知识服务存在的不足,联合双向长短期记忆网络(BiLSTW)、文本相似度等技术作为支撑,基于注意力机制的CNN的智能问答方式构建学术社区内资源的智能问答模型,增加注意力分支以抽取细粒度特征,提高检索速度,提升科研效率。结果/结论]智能化技术的应用实现学术资源的深度聚合,实现传统信息交互与智能化新技术的协同发展,提升了学术资源利用率,扩大信息资源共建共享。
|
| 关 键 词： | 虚拟学术社区 智能问答 深度学习 卷积神经网络 |
Research on Intelligent Question Answering in Virtual Academic Community Based on Deep Learning |
| |
| Xu Tongyang,Teng Qi.Research on Intelligent Question Answering in Virtual Academic Community Based on Deep Learning[J].Journal of Information,2021(4):163-169. |
| |
| Authors: | Xu Tongyang Teng Qi |
| |
| Institution: | (Faculty of Information,Shanxi University of Finance and Economics,Taiyuan 030006) |
| |
| Abstract: | Purpose/Significance]As an important place for researchers to acquire,exchange and utilize knowledge,virtual academic community resources are characterized by large quantity and high value.Exploring the theory,method and application of intelligent question-and-answer service in the community is of great significance for promoting the efficient retrieval and utilization of academic resources and promoting the innovation of knowledge service.Method/Process]The study combs the research status of virtual academic community and the shortcomings of current knowledge service,combined with bi-directional long-short-term memory network(BiLSTW),text similarity and other technologies as support,constructs an intelligent question-and-answer model of resources in academic community based on CNN’s intelligent question-and-answer mode of attention mechanism,and increases attention branches to extract fine-grained features,improve retrieval speed and enhance scientific research efficiency.Result/Conclusion]The application of intelligent technology realized the deep aggregation of academic resources,achieved the coordinated development of traditional information interaction and intelligent new technology,improved the utilization rate of academic resources,and expanded the co-construction and sharing of information resources. |
| |
| Keywords: | virtual academic community intelligent question and answer deep learning CNN |
| 本文献已被 维普 等数据库收录！ |
|
