Generalized Value Partition Problem: A Rough Set Approach

摘要

In this paper, the problem of partitioning purely qualitative, scaled qualitative, continuous numerical, and discrete numerical attribute values is generalized. A new algorithm for solving the topical problem of joint preprocessing of qualitative and quantitative attributes with missing values is proposed. An algorithm developed by the authors for obtaining the missing values is also presented. In the tests on databases from the UC Irvine Repository specially designed from real databases of various fields for testing and comparing generalization algorithms, the proposed algorithms have shown good mean classification accuracy equal to 90.4%.