Lei Zou (邹磊)

Professor

Wangxuan Institute of Computer Technology,

Peking University,No. 128 Zhongguancun North Street,

Haidian District, Beijing, 100871, P. R. China

Email: zoulei AT pku DOT edu DOT cn

I am currently a Professor at Wangxuan Institute of Computer Technology at Peking University and the deputy director of the National Engineering Research Center for New Electronic Publishing Technologies

I worked as an Associated Professor (Aug 2012 – Aug 2017), and as an Assistant Professor (Oct 2009 - Jul 2012) at Wangxuan Institute of Computer Technology, Peking University.

I visited Univerity of Waterloo (Oct. 2008-Oct.2009)，and visited Hong Kong University of Science and Technology (Dec. 2006-June. 2007). I received my B.Sc. Degree and Ph.D. degree in Computer Science at Huazhong University of Science and Technology (HUST) in 2003 and 2009, respectively.

My recent research interests include graph database, knowledge graph, hardware-accelerated graph computing systems and their applications in big data analysis. You may click the "Research" tab to know more about my research interests.

Positions Opening:

Our team (PKUMOD) are looking for excellent applicants: Research Scientist (Post doc position), Research Assistant Professor, Tenure-track Assistant/Associate Professor. Research directions, including but not limited to the following directions:

1) Database, especially graph database systems, graph query processing, etc.

2) Graph computing, including graph analysis, graph data mining, graph machine learning, and graph algorithm and graph computing-oriented hardware accelerator (GPU, FGPA, etc.).

3) Knowledge graph, including knowledge graph construction, quality control, and knowledge graph storage, retrieval, mining and application.

Please send a detailed CV to me including no more than 3 representative papers.

News

Apr 22,2024: One papers “LM-SRPQ: Efficiently Answering Regular Path Query in Streaming Graphs”was accepted by VLDB 2024
Feb 29,2024: One papers “Materialized View Selection & View-Based Query Planning for Regular Path Queries ”was accepted by SIGMOD 2024
Jan 25,2024: One papers “Rethinking and Extending the Probabilistic Inference Capacity of GNNs ”was accepted by ICLR 2024
Nov 7,2023: gStore was selected as BenchCouncil's Open100: Top 100 Open Source achievements；Lei Zou was selected as BenchCouncil's Top Open Source Contributors. Link:https://www.benchcouncil.org/evaluation/opencs/annual.html
Nov 1,2023: One papers “LLMaAA: Making Large Language Models as Active Annotators ”was accepted by EMNLP 2023
Aug 7,2023: Two papers accepted by CIKM 2023，“gFOV: A Full-Stack SPARQL Query Optimizer & Plan Visualizer ”and“CORD: A Three-Stage Coarse-to-Fine Framework for Relation Detection in Knowledge Base Question Answering”
May 5, 2023: Two papers accepted by ACL 2023，“AtTGen: Attribute Tree Generation for Real-World Attribute Joint Extraction”and “A Novel Table-to-Graph Generation Approach for Document-Level Joint Entity and Relation Extraction”
Apr 24, 2023: One paper“ Exploiting Ubiquitous Mentions for Document-Level Relation Extraction”was accepted by SIGIR 2023
Feb 28, 2023: One paper“FASI: FPGA-friendly Subgraph Isomorphism on Massive Graphs”was accepted by ICDE 2023
Feb 22, 2023: One paper“IFCA: Index-Free Community-Aware Reachability Processing Over Large Dynamic Graphs”was accepted by ICDE 2022
Feb 20, 2023: One paper“Sliding Window-based Approximate Triangle Counting with Bounded Memory Usage”was accepted by VLDBJ 2023
Dec 18, 2022: One paper“SGSI – A Scalable GPU-friendly Subgraph Isomorphism Algorithm”was accepted by TKDE 2022
Nov 27, 2022: One paper “gTop: An efficient SPARQL Query Engine” was accepted by APWEB-WAIM 2022 and won the Best Demo Paper Award.
Sep 30, 2022: Congratulate Bingcong Xue for the National Scholarship
Sep 20, 2022: One paper “GAMMA: A Graph Pattern Mining Framework for Large Graphs on GPU " has been accepted by ICDE 2023.
Sep 20, 2022: One paper “VEND: Vertex Encoding for Edge Nonexistence Determination " has been accepted by ICDE 2023.
Aug 20, 2022: Two papers accepted by ISWC, “VGStore: A Multimodal Extension to SPARQL for Querying RDF Scene Graph” and “Introducing Semantic Information for Numerical Attribute Prediction over Knowledge Graphs”
Jun 05, 2022: One paper “Graph Stream Sketch: Summarizing Graph Streams with High Speed and Accuracy” accepted by TKDE 2022.
Apr 11, 2022: One paper “Crake: Causal-Enhanced Table-Filler for Question Answering over Large Scale Knowledge Base”accepted by NAACL 2022.
Mar 24, 2022: One paper “Knowledge Graph Quality Management: a Comprehensive Survey” accepted by TKDE 2022.
Mar 23, 2022: One paper “Enhancing Chinese Pre-trained Language Model via Heterogeneous Linguistics Graph” accepted by ACL 2022.

Graph Data Management

Graph plays an important role in many big data applications due to the recent growth of social networks, RDF and other networks. Different from traditional RDBMS (relational database), graph database models the complex relations between different objects in an explicit way. Although the schema-free feature favors multi-source heterogeneous data integration, it brings lots of challenges to query optimization and distributed processing. My research focuses on building a native graph database system, including structure-aware index, query optimization mechanism and distributed graph query processing framework. Besides, we study optimizations of graph algorithms, including subgraph pattern match, path-related and random walk-based algorithms.

Knowledge Graph

Knowledge graph has attracted lots of attentions in the research community due to its prevalence in different applications. Generally, KG is an interdisciplinary research topic, since different researchers study KG from different perspectives. My research focuses on system-oriented topics: building a comprehensive KG system, including KG construction, KG data storage, management and applications, including KB-QA(question/answering). Specifically, we design a series of KG systems, such as gStore-an RDF graph database, gBuilder-a platform for KG construction and gAnswer-a KB Q/A system. Besides, we are currently studying multi-modal knowledge graph and reasoning.

Software/Hardware co-design for Graph Computing

Due to inherent hardness of graph problems, one desirable way to address this challenge is to employ hardware assist. However, different hardware-accelerators (such as GPU and FPGA) have different computation characters. My research is to design a series of quantitative models to consider both graph problems and hardware-accelerator computation features and model a hardware-accelerator design into an optimization problem. Besides, my research focuses on abstracting primitives among different graph algorithms and optimize these common computation patterns by the sophisticated software/hardware co-design strategy.

Current PhD students : Yue Pang, Xunbin Su, Yanzeng Li, Linglin Yang, Yinnian Lin, Lisheng Cao, Xinyi Ye，Sitan Ma, Chunshan Zhao, Jialun Zhong

Current master students: Ruoyu Zhang, Minhao Zhang, Tuo Xu, Yunfan Xiong, Zhiqiu Yuan

Current Postdoctororal: Siying Li

Graduated students

2024

Xiangyang Gou, Xinyi Ye, Lei Zou, Jeffrey Xu Yu: LM-SRPQ: Efficiently Answering Regular Path Query in Streaming Graphs. Proc. VLDB Endow. 17(5): 1047-1059 (2024)

Lei Zou (邹磊)

2022

2021​​

2020

2019

2018及以前

2021