宋 晖(教授)


宋 晖  

教授 研究生导师 信息技术系 系主任

本科、研究生就读于西安交通大学计算数学与软件专业,于上海交通大学获得计算机系统结构博士学位,曾赴美国纽约州立大学做访问学者。毕业后就职于太阳集团tcy8722,担任“智能科学与技术”专业负责人,全校计算机基础教学负责人。主要研究领域:自然语言处理、知识工程、智能信息系统。科研工作注重理论和实际应用相结合,研究成果在企事业单位中得到应用和推广,社会服务效应显著。多年来已培养研究生60余名。主持参与了国家自然科学基金项目3项、上海市重点科技攻关项目2项以及企业应用项目20余项,在国内外期刊、会议上发表研究论文40余篇,主持国家级一流本科课程1门,主编教材6本,获得国家专利发明3项。

主要社会任职:

中国图学应用图学专委会委员、中国人工智能学会会员、中国计算机学会会员、上海市计算机学会NLP专委会委员、上海市计算机学会会员、上海市计算机基础教育协会理事、上海市大学生计算机应用能力大赛评审委员会委员、国家高科技司评审专家

研究方向:

1. 面向领域的知识图谱构建、智能问答、大语言模型应用

2. 基于深度学习的知识挖掘

3. 智能信息系统、辅助决策

讲授课程:

研究生:  深度学习技术、概率统计与随机过程、数据挖掘(英语)

本科生:  自然语言处理、智能科学专业导论、数据科学技术与应用、Python语言程序设计

近期主持的科研项目:

1.面向知识推理的电力专业知识库构建与应用关键技术研究

2.基于多轮对话的电力行业问答系统关键技术研究

3.电力行业知识图谱构建及智能语义查询的关键技术研究

4.面向企业服务智能助手的关键技术研究与实现

5.基于知识获取的企业风控预警系统

6.政务领域软件机器人的研究与开发

7.基于大数据的智慧政务平台关键技术研究与开发

8.微信推广平台的开发与数据分析系统

9.电商与微博产品评论倾向性分析系统的研究与开发

10.中国民族民间音乐特征和噪声特点的知识数据库系统的研究与应用示范      

11.中国民族民间音乐知识库系统的研究与应用         

12.社会网络中新热点事件在线挖掘模型及关键技术的研究      

13.珍贵历史音频资料修复与数字化保存系统      

14.国家自然科学基金重大国际合作项目“中华民族音乐数字图书馆”      

15.Web信息抽取、检索系统模型,关键算法的设计与实现      

16.多策略的Web挖掘算法和模型的研究      

17.上海物流示范平台关键技术的研究      

18.基于ITILIT运维与监控系统       

19.分布式多Robot搜索引擎的研究与设计      

近几年的论著、专利:

1.Jie Yang, Hui Song, Bo Xu and Haoran Zhou. Dialogue State Tracking with a Dialogue-aware Slot-Level Schema Graph Approach. The 16th International Conference on Knowledge Science, Engineering and Management, KSEM 2023.

2.Bo Xu,Shizhou Huang,Ming Du,Hongya Wang,Hui Song,Yanghua Xiao,Xin Lin. A Unified Visual Prompt Tuning Framework with Mixture-of-Experts for Multimodal Information Extraction. The 28th International Conference on Database Systems for Advanced Applications, DASFAA 2023.

3.Shiyu Xu, Hui Song. A Natural Language Understanding Model Based on Encoding Fusion for Power Marketing Indicator Answering. Proceedings - 2023 2nd Asia Conference on Electrical, Power and Computer Engineering, EPCE 2023.

4.Bo Xu,Shouang Wei,Luyi Cheng,Shizhou Huang,Hui Song,Ming Du,Hongya Wang. HSimCSE: Improving Contrastive Learning of Unsupervised Sentence Representation with Adversarial Hard Positives and Dual Hard Negatives. IJCNN 2023.

5.MengMeng Gong, Haoran Zhou, Hui Song. Enhancing Response Relevance and Emotional Consistency for Dialogue Response Generation. 17th International Joint Symposium on Artificial Intelligence and Natural Language Processing, iSAI-NLP 2022 and 3rd International Conference on Artificial Intelligence and Internet of Things, AIoT 2022.

6.张洪宽, 宋晖, 徐波, 王舒怡, 基于BERT的端到端中文篇章事件抽取. 中文信息学报, 2022.12. (中文CCF B)

7.Xianglong He, Hui Song, Dong Cheng, Bo Xu, Few-Shot Relational Triple Extraction with Nearest Neighbor Matching. Proceedings of SPIE - The International Society for Optical Engineering, 2022.(EI)

8.Dong Cheng, Hui Song(B), Xianglong He, and Bo Xu. Joint Entity and Relation Extraction for Long Text, 14th International Conference of Knowledge Science, Engineering and Management, KSEM 2021. (CCF C, EI)

9.Shuyi Wang, Hui Song, Bo Xu, Hongkuan Zhang. Enhancing Embedding via Two-level Features for Machine Reading Comprehension. China Conference on Knowledge Graph and Semantic Computing. CCKS2020 (EI)

10.Longqi Dai, Bo Xu, and Hui Song. Feature-level Attention based Sentence Encoding for Neural Relation Extraction. The 8th CCF International Conference on Natural Language Processing and Chinese Computing. NLPCC2019. (CCF C, EI)

11.WanChen Yue, Xutong Liu, Hui Song. Analysis for Smart Urban Management Event Based on SARIMA and Spatiotemporal Clustering. Proceedings of the 2018 2nd international conference on Big Data Research. ICBDR 2018.EI

12.Xutong Liu, Hui Song, Bo Xu, Longqi Dai. Text analysis of cases description in urban management field based on knowledge atlas. Proceedings of 2nd International Conference on Computer Science and Application Engineering(ACM International Conference Proceeding Series), CSAE. 2018.EI

13.Xutong Liu, WanChenYue, Hui Song, Bo Xu. Text Analysis of Cases Description in Urban Management Field Based on Knowledge Atlas. 2018.

14.Chao Zhang, Hui Song, Zhenyu Liu, MiSAS: A Multi-domain Feature-level Sentiment Analysis System on Micro-blog. The 3rd International Conference on Communication and Information Processing Proceedings. 2017.(EI)

15.汪龙庆、张超、宋晖. 基于未登录词识别的微博评价短语抽取方法. 软件工程,Vol(34):6. 2017.

16.Dechao Zhu, Huisong, A Weighted Topical Document Embedding based Clustering Method for News Text, 2016 IEEE INFORMATION TECHNOLOGY, NETWORKING, ELECTRONIC AND AUTOMATION CONTROL CONFERENCE. 2016. (ISTP, EI)

17.Shan Chang, Ting Lu, Hui Song, “SmartDog: Real-time Detection of Smartphone Theft”, IEEE iThings 2016.

18.Yunjie Li, Hui Song, Applying Data Mining Techniques on Continuous Sensed Data for Daily Living Activity Recognition, Applied Mechanics & Materials, 2014, 738-739:191-196.

19.Hui Song, Jianfeng Chu, Semantic Analysis and Implicit Target Extraction of Comments from E-commerce Websites, The proceedings of 2013 4thWorld Congress on Software Engineering, 2013.

20.Hui Song, Botian Yang, Xiaoqiang Liu, Effective Comment Sentence Recognition for Feature-based Opinion Mining, 9th International Conference on Advanced Data Mining and Applications, LNAI 8347, 2013.

21.宋晖, 史南胜, 基于模式匹配与半监督学习的评价对象抽取, 计算机工程, Vol(39):10, P221-226, 2013.

22.Shan Chang, Hui Song, Li Lu, Qingsong Yao, Yong Qi, Random graph based benchmarking methodology for RFID security protocols, ICEBE2013, 2013.

23.发明专利,面向领域的中文文本主题句生成,2021

24.发明专利,一种检测和消除数字音频信号中脉冲噪声的方法. 2008.

25.发明专利,一种面向应用领域的软件质量基准测评方法. 2007.

联系方式:

电话:021-67792289

邮箱:songhui@dhu.edu.cn

办公室:上海市松江区人民北路29991号学院楼233


Hui Song

Professor, Head of Department of Information Technology

Research Areas:

1. Domain oriented knowledge map construction and intelligent question and answer

2. Knowledge mining based on deep learning

3. Intelligent information system, aided decision-making

Main Courses Taught:

Natural Language Processing, Data Science, Deep Learning technology, Probability Statistics and Stochastic Process, Data Mining

Publications:

1.Jie Yang, Hui Song, Bo Xu and Haoran Zhou. Dialogue State Tracking with a Dialogue-aware Slot-Level Schema Graph Approach. The 16th International Conference on Knowledge Science, Engineering and Management, KSEM 2023.

2.Bo Xu,Shizhou Huang,Ming Du,Hongya Wang,Hui Song,Yanghua Xiao,Xin Lin. A Unified Visual Prompt Tuning Framework with Mixture-of-Experts for Multimodal Information Extraction. The 28th International Conference on Database Systems for Advanced Applications, DASFAA 2023.

3.Shiyu Xu, Hui Song. A Natural Language Understanding Model Based on Encoding Fusion for Power Marketing Indicator Answering. Proceedings - 2023 2nd Asia Conference on Electrical, Power and Computer Engineering, EPCE 2023.

4.Bo Xu,Shouang Wei,Luyi Cheng,Shizhou Huang,Hui Song,Ming Du,Hongya Wang. HSimCSE: Improving Contrastive Learning of Unsupervised Sentence Representation with Adversarial Hard Positives and Dual Hard Negatives. IJCNN 2023.

5.MengMeng Gong, Haoran Zhou, Hui Song. Enhancing Response Relevance and Emotional Consistency for Dialogue Response Generation. 17th International Joint Symposium on Artificial Intelligence and Natural Language Processing, iSAI-NLP 2022 and 3rd International Conference on Artificial Intelligence and Internet of Things, AIoT 2022.

6.Hongkuan Zhang, Hui Song, Bo Xu, Shuyi Wang, Bert based Passage-level Event Extraction in Chinese Text. Chinese Journal of Information, 2022.12.

7.Xianglong He, Hui Song, Dong Cheng, Bo Xu, Few-Shot Relational Triple Extraction with Nearest Neighbor Matching. Proceedings of SPIE - The International Society for Optical Engineering, 2022.(EI)

8.Dong Cheng, Hui Song(B), Xianglong He, and Bo Xu. Joint Entity and Relation Extraction for Long Text, 14th International Conference of Knowledge Science, Engineering and Management, KSEM 2021. (CCF C, EI)

9.Shuyi Wang, Hui Song, Bo Xu, Hongkuan Zhang. Enhancing Embedding via Two-level Features for Machine Reading Comprehension. China Conference on Knowledge Graph and Semantic Computing. CCKS2020 (EI)

10.Longqi Dai, Bo Xu, and Hui Song. Feature-level Attention based Sentence Encoding for Neural Relation Extraction. The 8th CCF International Conference on Natural Language Processing and Chinese Computing. NLPCC2019. (CCF C, EI)

11.WanChen Yue, Xutong Liu, Hui Song. Analysis for Smart Urban Management Event Based on SARIMA and Spatiotemporal Clustering. Proceedings of the 2018 2nd international conference on Big Data Research. ICBDR 2018.EI

12.Xutong Liu, Hui Song, Bo Xu, Longqi Dai. Text analysis of cases description in urban management field based on knowledge atlas. Proceedings of 2nd International Conference on Computer Science and Application Engineering(ACM International Conference Proceeding Series), CSAE. 2018.EI

13.Xutong Liu, WanChenYue, Hui Song, Bo Xu. Text Analysis of Cases Description in Urban Management Field Based on Knowledge Atlas. 2018.

14.Chao Zhang, Hui Song, Zhenyu Liu, MiSAS: A Multi-domain Feature-level Sentiment Analysis System on Micro-blog. The 3rd International Conference on Communication and Information Processing Proceedings. 2017.(EI)

Main Research Projects:

1. Research on key technologies of power domain QA dialogue system

2. Research on key technologies of knowledge graph construction and intelligent semantic query in power domain

3. Research and implementation of key technologies of enterprise service oriented intelligent assistant

4. Enterprise risk control and warning system based on knowledge acquisition

5. Research and development of software robots in government affairs

6. Research and development of key technologies of smart government platform based on big data

7. Development and data analysis system of WeChat promotion platform

8. Research and development of e-commerce and microblog product review orientation analysis system

9. Research and application of knowledge base system of Chinese folk music

Address: Room 233, Building 1, North Renming Road, Songjiang District, Shanghai

Tel: 021-67792289

Mail: songhui@dhu.edu.cn


Baidu
sogou