Submissions are no longer being accepted. For published papers please see http://jmlr.csail.mit.edu/papers/topic/graphs_relations.html.
As data mining and machine learning techniques continue to evolve and improve, the role of structure in the data becomes more and more important. A major driving force is the explosive growth in the amount of heterogeneous data that is being collected in the business and scientific world. Early approaches to statistical learning were mainly based on vector-based data and attribute-value propositional representations. At the end of the 1990's, a "structured revolution" has started to profoundly change and extend the representational perspectives in all areas of machine learning and data mining. For example, the widespread diffusion of kernel methods has allowed several learning algorithms to abstract away data types and be applied to structured objects simply by plugging-in a suitable kernel function for the data type at hand. Yet, research has mainly focused on independent and identically-distributed (iid) examples. Dealing with inter-related examples that are linked together into complex graphs or hypergraphs remains one of the major challenges. Similarly, link and relation prediction, and supervised learning with structured outputs are substantially more difficult problems than single-output classification or regression.
Dealing with structured data has deep unresolved foundational and practical implications and affects different learning and mining paradigms. We therefore invite submission from research communities working on different theoretical and applicative aspects of machine learning and data mining, especially those that are active in cutting-edge frontier topics. These include, but are not limited to:
Application areas of interest are also diverse and include:
Submissions are no longer being accepted
A title and abstract must be sent by February 25th, 2008 to email@example.com.
The full manuscript must be submitted by March 3rd, 2008
using the JMLR submission system. Please follow the general JMLR author information when preparing your manuscript. Guest editors can be contacted at firstname.lastname@example.org.