Research Focus
大模型智能体
Our research aims to develop more capable, reliable, and efficient autonomous agents that can serve as general-purpose digital assistants across diverse real-world scenarios.
开放世界下的机器学习
Our research focuses on novel class detection and learning, few-shot class-incremental learning, transfer learning algorithms to handle challenging problems in open real-world scenarios.
图学习与推理
Our research focuses on key challenges in robustness, scalability, and knowledge-guided reasoning, aiming to empower intelligent systems with accurate and interpretable graph-based understanding.
交叉学科研究
Our research focuses on effectively integrating AI technologies with medicine, environmental data, and other real-world domains to address various challenges in the physical world.
DMLab Introduction
实验室致力于数据挖掘的基础理论及应用研究。团队目前包括教授、讲师、博士后、博士生、硕士生和优秀本科生组成的成员共50余人。其中教授2名，讲师1名，博士后2名，博士研究生16人，研究生30余名。团队成员入选国家级青年人才、德国洪堡学者、清华水木学者计划等，多次获四川省科技进步一等奖、二等奖。围绕大模型智能体、开放环境下的机器学习、可靠性数据流半监督学习、图神经网络、疾病AI可解释性辅助诊疗等领域,在KDD, ACL, ICLR, WWW, TKDE等CCF A及中科院一区论文发表大量论文, 其中五篇文章获最佳论文奖。在注重基础理论研究同时，面向国家重大战略需求，近年来承担了多项国家/省部级重点项目，主持的项目总经费超3000余万。包括国家重点研究计划课题，XXX基础加强重点项目课题、国家自然科学基金，XXX预研基金, 四川省杰出青年基金。同时，聚焦大数据行业领域共性技术问题，与华为建立长期预研合作（如华为AI锂电项目（515万）），5次获华为优秀项目奖和2次火花奖。
欢迎博士后、博士、研究生和优秀本科生加入，根据研究兴趣从事AI前沿研究和开发。联系人：邵俊明（junmshao@uestc.edu.cn）
-
The laboratory is dedicated to fundamental theories and applied research in the field of data mining. The team currently consists of more than 50 members, including professors, lecturers, postdoctoral researchers, Ph.D. students, master’s students, and outstanding undergraduate students. The team includes 2 professors, 1 lecturer, 2 postdoctoral researchers, 16 Ph.D. candidates, and more than 30 graduate students. The laboratory has undertaken more than 40 research projects, including National Key R&D Programs, key national defense projects, projects funded by the National Natural Science Foundation of China, and key R&D programs of Sichuan Province. Team members have been selected for prestigious talent programs such as the National Young Talent Program, the Alexander von Humboldt Foundation Scholars Program, and the Tsinghua University Shuimu Scholars Program. The team has also received multiple awards, including the First Prize and Second Prize of the Sichuan Provincial Science and Technology Progress Award. Focusing on key common technologies in industrial applications, the laboratory has established long-term research collaborations with companies such as Huawei, China Electronics Technology Group Corporation, and Siemens. The team has received the Huawei Outstanding Collaboration Award five times and the Huawei Spark Award twice.
Latest Works
见治澎同学一篇论文被地球科学与遥感领域顶级期刊IEEE TGRS接收
近日，电子科技大学数据挖掘实验室在地球科学与遥感领域国际顶级期刊《IEEE Transactions on Geoscience and Remote Sensing》（IEEE TGRS）上发表论文 “A Framework of Multisource Precipitation Data Fusion in the Yellow River Basin Based on Climate and Terrain Partitioning”。 该论文由见治澎（第一作者）、杨勤丽（通讯作者）、刘鸿源、邵俊明共同完成，电子科技大学为论文完成单位。
西轩志一篇广义类别发现论文被数据挖掘顶会KDD接收
Xuanzhi Xi, Zhong Zhang, Hongliang Wang, Tongze Zhang, Hongchao Hu, Qinli Yang, Junming Shao*. Topology-Aware Neural Collapse for Generalized Category Discovery on Graphs,32nd SIGKDD Conference on Knowledge Discovery and Data Mining (KDD, CCF A), 2026
李星昊一篇多源数据融合论文被水文顶刊JOH接收
Xinghao Li, Qinli Yang*, Junran Yang, Junming Shao. A Flow-Direction Guided Spatio-temporal Graph Neural Network (FD-STGNN) framework for multi-source soil moisture data fusion over the Qinghai-Tibet Plateau. Journal of hydrology, 2026
刘昕雨同学的一篇基于数据流演化的推荐算法被顶会ICDE接收
Xinyu Liu; Jinxia Guo, Qirui Hao, Zhongjing Yu, Qinli Yang, Junming Shao*. Community-level Personalized Recommendation by Exploiting Evolving User-Item Micro-clusters，IEEE International Conference on Data Engineering (ICDE, CCF A), 2026。
王红亮一篇数据流学习论文被人工智能领域会议IJCAI接收
Hongliang Wang, Hongyuan Liu, Qirui Hao, Yun Sing Koh, Qinli Yang, Junming Shao*. From Neural Collapse to Label-Limited Evolving Streams: Geometry-Constrained Learning under Dynamic Class Imbalance, 35th International Joint Conference on Artificial Intelligence (IJCAI), 2026
见治鹏一篇多源数据融合论文被IEEE TGRS接收
Zhipeng Jian, Qinli Yang*, Hongyuan Liu and Junming Shao. A framework of multi-source precipitation data fusion in the Yellow River Basin based on climate and terrain partitioning，IEEE Transactions on Geoscience and Remote Sensing (TGRS, IF=8.6), 2026
吴忠林一篇数据流半监督学习论文被中科院一区顶刊Information Fusion接收
Zhonglin Wu, Hongliang Wang, Tongze Zhang, Hongyuan Liu, Jinxia Guo, Qinli Yang, Junming Shao*. Region-based deep metric learning for tackling class overlap in online semi-supervised data stream classification, Information Fusion (IF = 15.5, 一区), 2026
张锐同学的一篇对样样本的论文被中科院一区ESWA期刊接收
Rui Zhang, Jörg Wicker, Katharina Dost, Qinli Yang, Zeyu Chen, Junming Shao*. Self-purification: Enhancing adversarial defense by leveraging local relative robustness,Expert Systems With Applications (IF = 7.5, 一区), 2026
王红亮同学一篇可靠性半监督学习论文被中科院一区期刊INS接收
Hongliang Wang; Zhonglin Wu; Jinxia Guo; Wei Han; Lei Liu; Qinli Yang; Junming Shao*.Exploiting Reliable Evolving Micro-clusters for Robust Semi-supervised Learning on Data Streams, Information Sciences (IF = 6.8, 一区), 2026