|
|
|
Topics of Interest
We aim at providing a forum for the discussion of recent advances in the use of machine learning and
data mining methods for learning monotone models, and at offering an opportunity for researchers and practitioners
to identify new promising research directions. Topics of interest include, but are not limited to:
- Monotone versions of popular classification and regression algorithms, for example: classification trees,
support vector machines, rule induction algorithms, Bayesian networks, neural networks, and nonparametric approaches.
- Empirical comparisons of monotone and non-monotone learning algorithms.
- Measures and tests to determine the degree of monotonicity of data sets.
- Partially monotone models, and models with different degrees of monotonicity.
- Discovery of monotone relationships.
- Theoretical foundations of learning monotone models.
- Applications of monotone models.
- Cleaning non-monotone data sets and generating monotone artificial data sets.
Papers on ordinal methods are welcomed as long as they also discuss monotonicity or order preservation.
Because of the scattered nature of research on monotonicity maintenance, different terminology has been used for comparable
problem areas such as supervised ranking and dominance based multi-criteria decision analysis, isotonic regression, isotonic
separation, order-restricted statistical inference, and so on. Needless to say that papers from such areas are welcomed.
|
|
