There is a great deal of interest in analyzing data that is best represented as a graph. Examples include the WWW, social networks, biological networks, communication networks, transportation networks, energy grids, and many others. These graphs are typically multi-modal, multi-relational, and dynamic. The importance of being able to effectively mine and learn from such data is growing, as more and more structured and semi-structured data is becoming available. Effectively learning from this kind of data poses several challenging problems, including:

• The necessity of a plethora of different techniques, including graph mining algorithms, graph embedding techniques, and other learning algorithms on graphs.
• Dealing with the heterogeneity of the data as well as information integration and alignment.
• Handling dynamic and changing data.
• Addressing each of these issues at scale.

Traditionally, a number of subareas have contributed to this space: communities in graph mining, learning from structured data, statistical relational learning, inductive logic programming, social network analysis, and network science. Our workshop will serve as a forum for researchers from this variety of fields working on mining and learning with graphs to share and discuss their latest findings.

The best paper was selected among the oral talks by the in-person audience of our workshop. The best poster was similarly selected by the audience among all posters presented at the workshop. We thank Astra Zeneca for the generous sponsorship of our two awards!

Best Paper Award

Removing Redundancy in Graph Neural Networks
by Franka Bause, Samir Moustafa, Wilfried Gansterer, Nils M Kriege

Best Poster Award

TAMARA: a task-aware multilayer graph simplification framework
Cheick Tidiane Ba, Roberto Interdonato, Dino Ienco, Sabrina Gaito

This workshop is a forum for exchanging ideas and methods for mining and learning with graphs, developing new common understandings of the problems at hand, sharing of data sets where applicable, and leveraging existing knowledge from different disciplines. The goal is to bring together researchers from academia and industry, to create a forum for discussing recent advances in graph analysis. In doing so, we aim to better understand the overarching principles and the limitations of current methods and to inspire research on new algorithms and techniques for mining and learning with graphs.

To reflect the broad scope of work on mining and learning with graphs, we encourage submissions that span the spectrum from theoretical analysis to algorithms and implementation, to applications and empirical studies. We are interested in the full spectrum of graph data, including but not limited to attributed graphs, labeled graphs, knowledge graphs, evolving graphs, transactional graph databases, etc.

We therefore invite submissions on theoretical aspects, algorithms and methods, and applications of the following (non-exhaustive) list of areas:

• Computational or statistical learning theory related to graphs
• Theoretical analysis of graph algorithms or models
• Semi-supervised learning, online learning, active learning, transductive inference, and transfer learning in the context of graphs
• Unsupervised learning and graph clustering
• Interesting pattern mining on graphs and community detection
• Graph kernels and metric learning on graphs
• Graph and vertex embeddings and representation learning on graphs
• Solving combinatorial problems on graphs with ML / data driven combinatorial optimization
• Explainable, fair, robust, and/or privacy preserving ML on graphs
• Statistical models of graphs and graph sampling
• Analysis of social media, chemical or biological networks, infrastructure networks, knowledge graphs
• Benchmarking aspects of graph based learning
• Libraries and tools for all of the above areas

We welcome many kinds of papers, such as, but not limited to:

• Novel research papers
• Demo papers
• Visionary papers (white papers)
• Appraisal papers of existing methods and tools (e.g., lessons learned)
• Relevant work that has been previously published
• Work that will be presented at the main conference (incl. research, ADS, or journal track; can be submitted in full length ignoring our page limit)

Authors must clearly indicate in their abstracts the kinds of submissions that the papers belong to, to help reviewers better understand their contributions. All papers will be (single blind) peer reviewed.

Submissions must be in PDF, long papers no more than 12 pages long, short papers no more than 8 pages long, formatted according to the standard Springer LNCS style required for ECMLPKDD submissions. Authors can use unlimited additional pages for references and appendices.

All accepted papers will be published on the workshop’s website. Unpublished submissions (e.g., novel research papers, novel demo papers, novel visionary papers, and novel appraisal papers) will additionally be considered for inclusion in the ECMLPKDD workshop proceedings in the Springer CCIS series. This is an opt-in process. Website publication will not be considered archival for resubmission purposes. Authors whose papers are accepted to the workshop will pitch their work to the full audience and will participate in a poster session. The best two submissions will also be chosen for a long oral presentation.

Please note that at least one author of each accepted paper has to register for the conference.

Dual Submission Policy: We accept submissions that are currently under review at other venues. However, in this case our page limits apply. Please also check the dual submission policy of the other venue.

For paper submission, please proceed to our submission page. When submitting a paper make sure to select the correct track “MLG: Mining and Learning with Graphs” from the list of workshops.

Please send enquiries to pascal.welke@tuwien.ac.at.

This year, the International Workshop on Mining and Learning with Graphs is collocated with two conferences. The ACM SIGKDD Conference on Knowledge Discovery and Data Mining, and the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases.

This allows researchers and practitioners from Europe and America to choose a venue that is geographically close and in a suitable time zone.

Feel free to visit the homepage of this year’s sister workshop: