The goal of GRADES-NDA is to bring together researchers from academia, industry, and government to create a forum for discussing recent advances in (large-scale) graph data management and graph analytics systems, as well as propose and discuss novel methods and techniques towards addressing domain-specific challenges in real-world graphs.

Proceedings

The proceedings of the workshop are published in the ACM Digital Library under the DOI 10.1145/3661304.

Keynote Speaker

We are very honored to host the following keynote speaker this year.

• Ioana Manolescu  Inria & Ecole Polytechnique
  Ioana Manolescu is a Senior Researcher at Inria, the national french institute of research in computer science and applied mathematics, and a part-time professor at Ecole Polytechnique, France's leading engineering school. Her main research interests are data integration, semistructured data management, and DB+AI methods and tools for data journalism and journalistic fact checking. She has co-authored 170 articles, several books, and is a recipient of the ACM SIGMOD 2020 Contribution Award. She has been a member of the SIGMOD Executive Committee and of the PVLDB Endowment Board, an Editor in chief of the SIGMOD Record, and a general chair of the ICDE conference.

Program

All times are in GMT-4 (the local time in Chile).

08:50 AM - 09:00 AM Welcome

09:00 AM - 10:00 AM Keynote: "The Real World is Graph-Structured: Retrieving Meaning from Heterogeneous Data" by Ioana Manolescu (Inria & Ecole Polytechnique) Abstract: Integrating heterogeneous data is an old problem with many successive twists. Challenging data integration problems often feature heterogeneity not only at the level of the schema, but also that of the data model. In the ConnectionLens project, we have shown that heterogeneous data of many semistructured models can be converted in fine-granularity graphs, which can be enriched and densified with the help of information extraction. Such fine-grain graphs, however, are hard to grasp for non-technical users. To help them get acquainted with a dataset, we devised an abstraction method, which identifies, in fine-granularity data graphs, structured objects endowed with an internal structure, and relationships between them. Our approach, implemented in a system called Abstra, inputs a semistructured dataset, and outputs a representation thereof resembling an Entity-Relationship dataset. In contrast with traditional E-R models, our entities may feature deep nesting, reflecting the nested and possibly recursive structure present in some data models. Thus, ConnectionLens with Abstra provide an automated way of "rescuing" the conceptual model, which we argue is best viewed as a graph, behind any application dataset. (Joint work with Nelly Barret, Prajna Upadhyay and many other colleagues) Speaker: Ioana Manolescu is a Senior Researcher at Inria, the national french institute of research in computer science and applied mathematics, and a part-time professor at Ecole Polytechnique, France's leading engineering school. Her main research interests are data integration, semistructured data management, and DB+AI methods and tools for data journalism and journalistic fact checking. She has co-authored 170 articles, several books, and is a recipient of the ACM SIGMOD 2020 Contribution Award. She has been a member of the SIGMOD Executive Committee and of the PVLDB Endowment Board, an Editor in chief of the SIGMOD Record, and a general chair of the ICDE conference.

10:30 AM - 10:45 AM Sponsor Talk 1

• Graph and Multi-Model in SAP HANA Matthias Hauck (SAP SE)

10:45 AM - 11:45 PM Paper Presentations

• Understanding High-Performance Subgraph Pattern Matching: A Systems Perspective Akshit Sharma, Dinesh Mehta, and Bo Wu
• Space & Time Efficient Leapfrog Triejoin Diego Arroyuelo, Daniela Campos, Adrián Gómez Brandón, Gonzalo Navarro, Carlos Rojas, and Domagoj Vrgoc

11:45 AM - 12:00 PM Sponsor Talk 2

• openCypher over RDF for Neptune Analytics Willem Broekema (Amazon)

02:00 PM - 03:30 PM Paper Presentations

• A Benchmark to Understand the Role of Knowledge Graphs on Large Language Model’s Accuracy for Question Answering on Enterprise SQL Databases Juan F. Sequeda, Dean Allemang, and Bryon Jacob
• HomeRun: A Cardinality Estimation Advisor for Graph Databases Wilco van Leeuwen, George H. L. Fletcher, and Nikolay Yakovets
• TelarKG: a Knowledge Graph of Chile's Constitutional Process Renzo Angles, Vicente Calisto, Javiera Díaz, Sebastián Ferrada, Aidan Hogan, Alexander Pinto, Juan Reutter, Carlos Rojas, Henry Rosales-Méndez, Hernan Sarmiento, Etienne Toussaint, and Domagoj Vrgoc

04:00 PM - 04:30 PM Paper Presentation

• Scaling Differential Computation for Large-Scale Graph Processing Siddhartha Sahu and Semih Salihoglu

04:30 PM - 05:30 PM Invited Talk: "PODS Goes Practical: Can Database Theory be of any use for Building Graph Systems?" by Domagoj Vrgoč (Pontificia Universidad Católica de Chile) Abstract: In this talk we will present our experience of what it takes to implement core theoretical findings inside of a working graph database engine. To do so, we will present how MillenniumDB, the engine built at IMFD Chile, uses the idea of worst-case optimal algorithms and automata-based path search to efficiently evaluate important classes of graph queries. The presentation will focus on the main challenges that an implementer faces when adapting theoretical work and try to give some insight on how these approaches fit into a classical database architecture, or how to adapt them for alternative storage models. During the presentation we will try to give a wider overview of the class of enumeration algorithms and discuss why these might be a good theoretical framework for algorithms that are implementable and have the potential to perform well in practice.

Program Committee

• Renzo Angles, Universidad de Talca
• Marcelo Arenas, PUC Chile
• Amitabha Bagchi, Indian Institute of Technology, Delhi
• Kaustubh Beedkar, IIT Delhi
• Yang Cao, The University of Edinburgh
• Nathalie Charbel, Neo4j
• Juan Colmenares, Microsoft
• Sourav Dutta, Huawei Research
• George H. L. Fletcher, Eindhoven University of Technology
• Russ Harmer, CNRS & ENS Lyon
• Jan Hidders, Birkbeck College, University of London
• Davide Mottin, Aarhus University
• Nikos Ntarmos, Huawei Technologies R&D (UK) Ltd
• Evaggelia Pitoura, Univ. of Ioannina
• Petra Selmer, Bloomberg
• Marco Serafini, University of Massachusetts Amherst
• Hiroaki Shiokawa, University of Tsukuba
• Vasileios Trigonakis, Oracle Labs
• Hannes Voigt, Neo4j
• Yinghui Wu, Case Western Reserve University
• Yinglong Xia, Facebook
• Yuichi Yoshida, National Institute of Informatics
• Shangdi Yu, MIT

Sponsored by