Angelov, P.P. and Gu, X. (2019) Applications of Autonomous Data Partitioning. In: Empirical Approach to Machine Learning :. Studies in Computational Intelligence, 800 . Springer-Verlag, pp. 261-276. ISBN 9783030023836)
Full text not available from this repository.Abstract
In this chapter, the algorithm summaries of both, the offline and evolving versions of the proposed autonomous data partitioning (ADP) algorithm described in chapter 7 are provided. Numerical examples based on well-known benchmark datasets are presented for evaluating the performance of the ADP algorithm on data partitioning. Furthermore, numerical examples on semi-supervised classification are also conducted as a potential application of the ADP algorithm. The state-of-the-art approaches are used for comparison. Numerical experiments demonstrate that the ADP algorithm is able to perform high quality data partitioning results in a highly efficient, objective manner. The ADP algorithm can also be used for classification even when there is very little supervision available. The pseudo-code of the main procedure of the ADP algorithm and the MATLAB implementations can be found in appendices B.2 and C.2, respectively. © 2019, Springer Nature Switzerland AG.