Workshop on

Supervised and Unsupervised

Ensemble Methods and Their Applications

 

(in conjunction with IbPRIA’2007)

Ensembles of supervised learning machines and, in particular, ensembles of classifiers have been established as one of the main research topics in machine learning. Methods for combining unsupervised clusterings have been proposed to improve the reliability and to assess the validity of discovered clusters. The main goal of this workshop is to provide a forum open to researchers from pattern recognition and related disciplines to present and discuss problems related to unsupervised and supervised ensemble methods with a particular focus on their applications to real-world problems. In particular, several problems remain open: for instance, the search of the “best” set of base classifiers or the “best” set of combination methods with respect to the characteristics and the distribution of the data; theoretical reasons of the practical success of several widely used ensemble methods; relationships between the diversity and accuracy of base classifiers forming an ensemble. Moreover, ensembles of classifiers have been successfully applied for a large set of real-world classification tasks, and recently, innovative applications in the field of unsupervised learning have been proposed. This workshop intends to provide a forum for scientists from and outside the Iberian Peninsula. Possible topics of the workshop include (but are not limited to):

• New ensemble methods raised from new real world supervised and unsupervised learning problems.

• Application of ensemble methods in various branches of science and technology with a particular focus on:

– bioinformatics,

– computer security,

– medical informatics,

– ecology,

– economics,

– meteorology and weather forecast,

– satellite image analysis.

• Fusion of multiple-source/multi-sensor data.

• Unsupervised ensemble methods for discovering structures in unlabeled real data.

• Unsupervised ensemble approaches to assess the reliability/validity of clusters discovered in real data.

• Combination techniques and methods to generate multiple base learners from different features and data.

• Classifier selection strategies for ensemble.

• Heterogeneous ensembles of base learners.

• Variants of resampling-based methods (bagging, boosting).

• Ensemble methods for supervised multi-class classification and regression.