Dynamic updating approximations of local generalized multigranulation neighborhood rough set

The approximation space in rough set theory is important for dealing with uncertainties. As the information contained in various information systems is constantly updated and changed with the development of information technology, how to effectively obtain the approximation space in dynamic environments is essential. The local rough set as an excellent model avoids unnecessary calculation of information granules, and can significantly improve learning efficiency. In this paper, we mainly investigate a dynamic approximation update mechanism of multigranulation data from local viewpoint. We first define a support and inclusion function to construct local generalized multigranulation neighborhood rough set model. Then, the dynamic updating process of global rough set and local rough set is analyzed when object chandes. Meanwhile, the corresponding dynamic update algorithms for dynamic objects are proposed based on local generalized multigranulation rough set model. The complexity analysis about them theoretically proves the efficiency of local dynamic algorithm compared with global algorithm and static algorithm. To illustrate the effectiveness of proposed algorithms, twelve datasets from UCI are adopted to contrast experiments.

Automated summary

The approximation space in rough set theory is crucial for handling uncertainties. The local rough set model, as an effective approach, improves learning efficiency by avoiding unnecessary information granule calculations. This paper investigates the dynamic approximation update mechanism for multigranulation data and proposes corresponding dynamic update algorithms based on the local generalized multigranulation rough set model.