Shenghua Liu

Associate Professor (short bio)

Institute of Computing Technology (ICT)

Chinese Academy of Sciences (CAS)

  • Email:  liushenghua at ict.ac.cn
  • Address:  NO. 6, Kexueyuan South Road, Haidian, Beijing, P.R.China, 100190
  • Biography
  • Publication
  • Softwares
  • Students

He is now an Associate Professor at Institute of Computing Technology, Chinese Academy of Sciences.

He spent his one-year sabbatical at Computer Science Department, Carnegie Mellon University (CMU), as a research scholar. He was hosted and supervised by Professor Christos Faloutsos, 2016-2017.

He received his Ph.D. degree from Computer Science & Technology Department, Tsinghua University in 2010, and was funded by hosting Professor Lei He to visit Electronic Engineering Department, University of California, Los Angeles (UCLA) as a Ph.D. student, 2006-2007. He is in consequence listed as one of the Alumni in Academia of Electrical & Computer Engineering, UCLA.

He focuses on research areas such as big graphs and time series analysis, large language models, and scalable algorithm design. His expertise is applied to interpreting complex patterns across diverse domains, including online user behavior within social platforms and e-commerce, as well as various networked systems ranging from academic collaborations, supply chain, financial transactions and biological networks. The featured works are published on IEEE TKDE, ACM TKDD, and proceedings of top-tier conferences such as AAAI, CIKM, WSDM, IJCAI, ECML-PKDD, ICDM, and SDM. Some of the publications are recognized as ASP-DAC 2010 best paper candicate, ECML-PKDD 2020 best student DM paper award.

NEWS

Honors and Awards
  • Outstanding research staff, Institute of Computing Technology, Chinese Academy of Sciences, 2012, 2018
  • Outstanding dissertation, Tsinghua University, 2010. [ slides ][ demo1, demo2 ]
  • Microsoft Research Asia (MSRA) fellowship, 2009.
  • The honor of Academic Rising Star, Dept. of CS & T, Tsinghua Univisity 2009
Academic Services
  • One of guest editors, Complexity, Special Issue: Deep Structure Representation and Learning for Complex Information Networks, deadline 09 Oct 2020.
  • Reviewer of IEEE Transactions on Knowledge and Data Engineering(TKDE), ACM Transactions on Knowledge Discovery from Data(TKDD), ACM Transactions on Intelligent Systems and Technology (ACM TIST), Chinese Journal of Computers(JSC), Chinese Journal of Computers, etc.
  • PC member of The AAAI Conference on Artificial Intelligence (2016- ).
  • Committee member of Social Media Processing of Chinese Information Processing, (2013- )
Representative publications

    Topological and geometric graph-based algorithms in application of large-scale circuit design

  • Shenghua Liu, Guoqiang Chen, Tom Tong Jing, Lei He, Tianpei Zhang, Robi Dutta, Xian-Long Hong, Topological Routing to Maximize Routability for Package Substrate, In Proc. of 45th ACM/EDAC/IEEE Design Automation Conference (DAC), Anaheim, June 2008, pp.566-569. [ paper ]
  • Shenghua Liu, Yuchun Ma, Xian-Long Hong, Yu Wang, Simultaneous Slack Budgeting and Retiming for Synchronous Circuits Optimization, In Proc. of 15th Asia and South Pacific Design Automation Conference (ASP-DAC), 2010, pp. 49-54. (Best paper candidate) [ paper ]
  • Shenghua Liu, Guoqiang Chen, Tom Tong Jing, Lei He, Robi Dutta, Xian-Long Hong, Diffusion-Driven Congestion Reduction for Substrate Topological Routing, In Proc. of International Symposium on Physical Design (ISPD SIGDA) 2009, Mar. 2009, pp175-180. [ paper ]
  • Shenghua Liu, Guoqiang Chen, Tom Tong Jing, Lei He, Tianpei Zhang, Robi Dutta, Xian-Long Hong, Substrate Topological Routing for High Density Packages, IEEE trans. on Computer Aided Design (TCAD), Vol.28, No.2, 2009, pp207-216. [ paper ]
  • Shenghua Liu, Guoqiang Chen, Tom Tong Jing, Lei He, Robi Dutta, Xian-Long Hong, Effective Congestion Reduction for IC Package Substrate Routing, ACM Trans. on Design Automation of Electronic System (TODAES), Vol. 15, Issue 3, 2010. [ paper ]
Dissertation (Chinese)
2024
2023
2022
2021
2020
  • [C] Yiwei Wang, Shenghua Liu, Minji Yoon, Hemank Lamba, Wei Wang, Christos Faloutsos, and Bryan Hooi, Provably Robust Node Classification via Low-Pass Message Passing, In proc. of IEEE International Conference on Data Mining (ICDM), Sorrento, Italy, November 17-20, 2020. ( acceptance rate 9.8% for regular papers. Theoretically upper-bounded under adversarialy attacks; Easy-to-plugin module for graph neural networks; As robust as in linear of attack budget, and as accurate as neural networks. ) [ slides ]
  • [C] Wenjie Feng, Shenghua Liu, Danai Koutra, Huawei Shen, Xueqi Cheng, " SpecGreedy: Unified Dense Subgraph Detection ", In proc. of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), Belgium, Sept 2020. (Best student DM paper award. Acceptance rate of 19%. Verified on 40 real-world networks, and a 1.47-billion-edge graph )
  • [J] Wenjie Feng, Shenghua Liu, Christos Faloutsos, Bryan Hooi, Huawei Shen, and Xueqi Cheng. EagleMine: Vision-guided Micro-clusters recognition and collective anomaly detection , Future Generation Computer Systems, Vol 115, Feb 2021, pp.236-250.
  • [C] Xiangfeng Li, Shenghua Liu, Zifeng Li, Xiaotian Han, Chuan Shi, Bryan Hooi, He Huang, Xueqi Cheng. " FlowScope: Spotting Money Laundering Based on Graphs ," Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI) 2020. [ appendix ] [ code ] [ slides ]
2019
2018
2017
2016
  • [C] Tong Man, Huawei Shen, Shenghua Liu, Xiaolong Jin and Xueqi Cheng, Predict Anchor Links across Social Networks via an Embedding Approach, In Proc. of the 25th International Joint Conference on Artificial Intelligence IJCAI-16, pp. 1823-1829, July 9-15, New York City, New York, USA.
  • [O] Shenghua Liu and Xueqi Cheng, Social Media Sentiment Analysis: opinions, users, and behaviors, in Communication of Chinese Association for Artificial Intelligence (CAAI), Special Issue on Social Media Analysis, 2016:12-17.
  • [C] Jinhua Gao, Huawei Shen, Shenghua Liu, Xueqi Cheng, Modeling and predicting retweeting dynamics via a mixture process, In Proc. of the 25th international conference on World Wide Web (WWW) (Companion Volume), 2016. [link]
2015
2014
2013
2012
2011
until 2010
[Patents]
  • DYNAMIC PUSH FOR TOPOLOGICAL ROUTING OF SEMICONDUCTOR PACKAGES, filing date: June 6, 2008, patent number: 8006216, issue date: Aug. 23, 2011, Guoqiang Chen, Kaushik Sheth, Egino Sarto, and Shenghua Liu.(US patent, issued)
  • A FAST ROUTABILITY ESTIMATING METHOD IN VLSI, filing date 6/15/2006 (ZL 200610012271.4), issued date 7/23/2008 (CN 100405379C), by Xianlong Hong, Tong Jing, Shenghua Liu, and Jingyu Xu. (Chinese patent, issued)
  • AN AUTOMATED SOCIAL TAGGING METHOD AND SOCIAL TAG GENERATOR, filing date: 9/7/2011, patent number 201110263798.5, Shenghua Liu, Xueqi Cheng, Jiafeng Guo, Yue Liu, Huaming Liao, and Yatao Zhu.(Chinese patent, filed)
  • AN OPEN-KNOWLEDGE-BASED SEMANTIC CONCEPT LINKING AND EXTENSION METHOD FOR SHORT TEXT, filing date: 3/15/2013, patent number 201310081984.6, issue date: 4/1/2015, issue number: CN 103150382B, Xueqi Cheng, Shenghua Liu, Yonglei Xiao, Yuanzhuo Wang, and Yue Liu. (Chinese patent, issued)
  • AN EVENT EVOLVEMENT ANALYSING METHOD FOR SHORT TEXT, filing date: 3/15/2013, patent number: ZL 201310082990.3, Xueqi Cheng, Shenghua Liu, Fuxin Li, Yuanzhuo Wang, and Yue Liu. (Chinese patent, issued)
  • A VISUALIZING METHOD FOR DYNAMIC OPINIONS IN SOCIAL MEDIA, file date: 4/18/2013, patent number: ZL 201310134433.1, Xueqi Cheng, Shenghua Liu, Yatao Zhu, Yuanzhuo Wang, Yue Liu, and Wenjun Zhu. (Chinese patent, issued)
  • A REALTIME EVENT FILTERING METHOD AND SYSTEM FOR STREAMING WEB DATA, filing date: 4/19/2013, patent number: ZL 201310136896.1, Xueqi Cheng, Shenghua Liu, Wenyi Qiu, Yuanzhuo Wang, Yue Liu, Yi Mo, and Zhankun Huang. (Chinese patent, issued)
  • spartan2: a developing open-sourced graph and time series mining package based on sparse tensor/matrix and sequential analysis. [git repository]
  • HoloScope:Topology-and-Spike Aware Fraud Detection in big graph [ code ]
  • SpecGreedy: Unified Dense Subgraph Detection [code]
  • Heatmap for point list [ download ]
  • NeuCast for time series forecasting in power grid [download]
  • BeatGAN: Anomaly detection in time series, e.g. ECG, and sensor data of motions. [ code ]
  • EagleMine: vision-guided anomaly detection in large graphs [ code ]
  • FlowScope: fast algorithm for multipartite subgraph detection, and used for spotting money laundering [ code ]
  • EigenPulse: detect anomalies in streaming graphs [code]
  • CatchCore: detecting dense blocks which show hierachical structures and have a core. [ code ]

Disseration demo

Students under supervision
  • Yuyao Ge (Ph.D. candidate, advising on large language models and graphs, 2023 (senior))
  • Xuanshan Zhou (Master candidate, advising on scalable graph summarization, 2023 (senior))
  • Baolong Bi (Ph.D. student, advising on foundation model and graphs, 2022(senior), 2023.9- )
  • Tianjie Hou (Master student, advising on graph anomaly detection, 2022(senior), 2023.9- )
  • Li Wang (Master student, advising on temporal graph mining, 2021(senior), 2022.9- )
  • Lingrui Mei (Master student, co-advising on large language models, (2022) 2023.9- )
  • Xiaobing Sun (Master student, advising on anomaly detection and graph learning, 2020(senior), 2021.9- )
  • Houquan Zhou (Ph.D. student, co-advising on big graph summarization and representation learning, 2018.9- )
Alumni
  • Yugao Zhu: advised on big graph mining algorithm and spectral graph theory 2021-2023, transferred to HKUST since 2024
  • Caizheng Liu: Ph.D., co-advised on time series mining, 2017 - 2023.6, now a post-doc
  • Quan Ding: Master, advised on time series mining and anomaly detection, 2019(senior), 2020.9-2023.6, now at a startup, Shenzhen
  • Siwei Zeng: Master, advised on big graph mining, 2018(senior), 2019.9-2022.6, now at ByteDance, Beijiing
  • Bowei Lin: Master, co-advised on graph matching, 2019.9-2022.6, now a R&D Engineer at ICT
  • Jiabao Zhang: Master, advised on streaming graph and tensor mining, 2017(senior), 2018.9-2021.6, now at Didi research group
  • Wenjie Feng : co-advised Ph.D. student, big graph mining, 2016 - 2020.9, now Post-doc at NUS, Singapore
  • Bin Zhou: Master, advised on time series mining, 2016 (senior), 2017.9-2020.6, now at Meituan.
  • Haiyin Zhang: visiting undergraduate, 2019.4-2020.6, now Master student at TU Delft, Dutch.
  • Xin Zhao (undergraduate, co-advised on big sparse tensor computing system, 2018(senior) )
  • Xiangfeng Li (co-advised graduate student, visiting, 2018.5-2020.6)
  • Xiaotong Jiang (visiting undergraduate, 2018.7-2019.7, now Master student at USC, US )
  • Pudi Chen (co-advised graduate student, visiting, 2017.10-2019.6, now at Ant financial services group)
  • Yang Liu (visiting Master, 2016-2017)
  • Qi Cao (co-advised Master, now Ph.D. at ICT CAS, 2017-)
  • Yongqing Wang (co-advised Ph.D., 2015-2017, now associate professor at ICT)
  • Tong Man (co-advised Ph.D., now at Amozon U.S.)
  • Yonglei Xiao (Master, now at Sohu)
  • Houdong Zheng (visiting Master, Fall 2015 - Fall 2016)
  • Yatao Zhu (Master, now ph.D.)
  • Yi Mo (Master, now at Tencent)
  • Zhankun Huang (Master, now at Netease)
  • Peng Cao (co-advised Ph.D. now overseas)
  • Fuxin Li (Master, now at MI Technology)
  • Wenyi Qiu (visiting undergraduate from CUG, now at ZhongRuan Inc)
  • Bin Wang (visiting Master from WHUT)
  • Wenjun Zhu (visiting Master from WHUT)
  • Xiaoli Wang (visiting Master from Xidian Univ.)
  • Wenjing Zhao (visiting Master from Xidian Univ.)