Rseslib: Rough Sets, Rule Induction and Analogy-Based Reasoning

The package provides 4 classifiers. The rule classifier RoughSet uses the concepts of discernibility matrix, reducts and rules generated from reducts. It provides variety of algorithms generating reducts including giving more general rules local reducts and has modes to work with incomplete data and inconsistent data. The k nearest neighbors classifier RseslibKnn provides variety of distance measures that can work also for data with both numeric and nominal attributes and has built-in k optimization. It implements a fast neighbors searching algorithm making the classifier work for very large data sets. The classifier has also the mode to work as RIONA algorithm. The LocalKnn classifier is the extension of the k nearest neighbors method that induces a local metric for each classified object. It is dedicated rather to large data sets (2000+ training instances) and improves accuracy particularly in case of data containing nominal attributes. The RIONIDA classifier dedicated to imbalanced data with two decision classes combines instance-based learning with rule induction. It enables to differentiate the importance of the decisions and to control the impact of rules on the decision selection process and applies multi-dimensional optimization of classification measures relevant for imbalanced data.

All available versions:
Latest
3.3.1
3.3.0
3.2.5
3.2.4
3.2.3
3.2.2
3.2.1
3.2.0
3.1.2
3.1.1
3.1.0