Rough Sets-Based Machine Learning over Non-deterministic Data: A Brief Survey

Abstract

Rough Non-deterministic Information Analysis (RNIA) is a rough sets-based framework for handling tables with exact and inexact data. Under this framework, we investigated possible equivalence relations, data dependencies, rule generation, rule stability, question-answering systems, as well as missing and interval values as special cases of non-deterministic values. In this paper, we briefly survey RNIA, and report the state of its underlying software implementation. We also discuss to what extent RNIA can be seen as an example of a new emerging paradigm in machine learning. © Springer-Verlag Berlin Heidelberg 2012.