Rseslib: Rough Sets and Analogy-Based Reasoning

Author:Arkadiusz Wojna,Grzegorz Gora,Marcin Jalmuzna,Michal Kurzydlowski,Rafal Latkowski,Dariusz Ogorek,Marcin Piliszczuk,Beata Zielosko
Maintainer:Arkadiusz Wojna <wojna{[at]}>

The package provides 3 classifiers. Rule classifier based on rough sets uses the concepts of discernibility matrix, reducts and rules generated from reducts. It provides variety of algorithms generating reducts including faster for larger data sets local reducts and has modes to work with incomplete data and inconsistent data. K nearest neighbours classifier 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 fast neighbours searching algorithm making the classifier work for very large data sets. The classifier has also the mode to work as RIONA algorithm. The third classifier is the extension of k nearest neighbours method that induces 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.

All available versions: