Call for Papers
Scope
Ordinal regression (so called ranking, sorting or ordinal classification) is a relatively new learning problem, where the objective is to learn a rule to predict labels in an ordinal scale, the labels being discrete but being possible to stablish a natural order among them. Consider, for example, a teacher who rates student performance using A, B, C, D and E and we know that A>B>C>D>E.
Many real problems require the classification of items into naturally ordered classes, e.g. Multi-criteria decision making, Medicine, Risk analysis, University ranking, Information retrieval and filtering, and, in general, problems involving humans participating in the data generation process.
Ordinal regression has been commonly tackled as standard multinomial classification, ignoring ordering information, and penalising equally all mistakes. Others have considered ordinal regression as standard regression problems, assigning a numerical value for each class, what is difficult and very problem-dependent. Alternatively, some specific solutions have been recently proposed in Machine Learning and Pattern Recognition literature, ordinal regression being a very active and interesting field. This special session aims to cover a wide range of works and recent advances on ordinal regression. We hope that this session can provide a common forum for researchers and practitioners to exchange their ideas and report their latest finding in the area.
Topics
Original contributions are solicited in the following topics (but are not limited to):
- Goodness-of-fit Tests for Ordinal Response Regression Models.
- Evaluation Measures for Ordinal Regression.
- Imbalanced Ordinal Regression Problems.
- Ordinal regressión by extreme learning machines.
- Latent variable model for categorical data.
- Generalized linear models with ordered predictors.
- Multivariate Analysis of Ordinal Measures.
- Threshold models for ordinal discrete data.
- Proportional odd logistic regression analysis of ordinal score data.
- ROC analysis in ordinal regression learning.
- Modelling ordinal relations with SVMs.
- Information entropy for Ordinal Regression.
- Adding monotonicity to learning algorithms.
- Distribution-based models for the classification of ordinal data.
- Ordinal versus nominal classification.
- Probabilistic kernel approach to ordinal regression based on Gaussian processes.
- Ranking, reranking, and ordinal regression algorithms.
- Replicating data for ordinal regression.
- Kernel Discriminant Learning for Ordinal Regression.
- Conditional risk models for ordinal response data.
- Soft Computing for ordinal regression.
- Bioinspired algorithms for ordinal regression.
- Data Mining algorithms for ordinal regression.
- Evolutionary algorithms for ordinal regression.
- Applications in Medicine, Information Retrieval, Risk Analysis… and any other real problems involving a set of ordered labels.
Paper Submission
Please follow the instructions given at the corresponding section. If you are thinking about submitting a paper, please send an email to the organizers as soon as possible, with the title of the work and the authors.
Organizers
- Pedro Antonio Gutiérrez: pagutierrez@uco.es
Department of Computer Science and Numerical Analysis, University of Córdoba, Spain. - Francisco J. Martínez-Estudillo: fjmestud@etea.com
Department of Management and Quantitative Methods, ETEA, University of Córdoba, Spain. - César Hervás Martínez: chervas@uco.es
Department of Computer Science and Numerical Analysis, University of Córdoba, Spain.