MLAG 2014 - Workshop on Mining and Learning of Augmented Graphs

Graphs are a powerful mathematical abstraction that enable to depict and study many real-word phenomena. Mining and learning graphs have been extensively studied for than two decades. As a result, data mining and machine learning are now sufficiently mature to handle large graphs. With the rapid development of social media, sensor technologies and bioinformatic assay tools, real-world graph data has become ubiquitous. In practice, the simple graph structure is not enough to catch all the complexity/diversity of phenomena depicted by the graphs. Indeed, many pieces of information are often available and can be used to produce augmented graphs such as multidimensional graphs, attributed graphs, dynamic graphs, multi-level graphs, 3D-graphs, etc. It is a timely challenge to provide some data mining and machine learning methods to analyze such augmented graphs.
The aim of this workshop is to bring together pioneering researchers in the fields of data mining and machine learning which are focusing in their research on the analysis of the non trivial aspects of augmented graphs. This workshop is a forum for exchanging ideas and methods for mining and learning of augmented graphs, developing new common understandings of the problems at hand, sharing of data sets where applicable, and leveraging existing knowledge from different disciplines. In doing so we aim to better understand the overarching principles and the limitations of our current methods, and to inspire research on new algorithms and techniques for mining and learning on augmented graphs.
Hence, the topics of interest include, but are not limited to:
Theoretical aspects:
Mining and learning in augmented graphs;
Modeling and visualizing of augmented graphs;
Constraint-based pattern mining in augmented graphs;
Privacy preserving mining of augmented graphs;
Data mining query languages for augmented graphs;
Computational or statistical learning theory related to augmented graphs;
Analysis of heterogeneous network;
Analysis of attributed graphs;
Analysis of multidimensional graphs;
Analysis of 3D-graphs;
Relationships between augmented graphs and statistical relational learning or inductive logic programming
Analysis of complex systems;
Methods:
Graph mining algorithms;
Graphs kernel algorithms;
Relational learning algorithms;
Matrix/Tensor methods;
Bayesian methods;
Clustering/Co-clustering/Biclustering;
Subspace clustering;
Pattern mining with constraints;
Subgroup discovery;
Community detection;
Applications:
Social networks;
Social media;
System biology;
Sensor networks;
Our main goal is to stimulate discussion, collaboration and the sharing of experiences. In that respect, we have three submission types:
unpublished works (max 15 pages, double submissions allowed). We allow a submitted or under review paper to also be submitted to the workshop. In this way, we offer authors reviews and (if accepted) discussion on their work among workshop participants.
extended abstracts and vision statements (max 6 pages). Short papers and vision statements are meant to be thought provoking and stimulate discussion.
recently published works (special oral-only track, no page limits) Part of the program will be for a short oral-only presentation of recently published work. The aim is to share experiences and lessons learned
The submission system can be reached through this link (easychair link). Submissions (except oral-only) should follow the LNCS formatting guidelines.