Multi-Objective Evolutionary Instance Selection for Regression Tasks
The purpose of instance selection is to reduce the data size while preserving as much useful information stored in the data as possible and detecting and removing the erroneous and redundant information. In this work, we analyze instance selection in regression tasks and apply the NSGA-II multi-obje...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2018-09-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1099-4300/20/10/746 |
| _version_ | 1798000041944678400 |
|---|---|
| author | Mirosław Kordos Krystian Łapa |
| author_facet | Mirosław Kordos Krystian Łapa |
| author_sort | Mirosław Kordos |
| collection | DOAJ |
| description | The purpose of instance selection is to reduce the data size while preserving as much useful information stored in the data as possible and detecting and removing the erroneous and redundant information. In this work, we analyze instance selection in regression tasks and apply the NSGA-II multi-objective evolutionary algorithm to direct the search for the optimal subset of the training dataset and the k-NN algorithm for evaluating the solutions during the selection process. A key advantage of the method is obtaining a pool of solutions situated on the Pareto front, where each of them is the best for certain RMSE-compression balance. We discuss different parameters of the process and their influence on the results and put special efforts to reducing the computational complexity of our approach. The experimental evaluation proves that the proposed method achieves good performance in terms of minimization of prediction error and minimization of dataset size. |
| first_indexed | 2024-04-11T11:13:57Z |
| format | Article |
| id | doaj.art-5a26340f4fec4ebaa64d2beb0bd64f0f |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-04-11T11:13:57Z |
| publishDate | 2018-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-5a26340f4fec4ebaa64d2beb0bd64f0f2022-12-22T04:27:18ZengMDPI AGEntropy1099-43002018-09-01201074610.3390/e20100746e20100746Multi-Objective Evolutionary Instance Selection for Regression TasksMirosław Kordos0Krystian Łapa1Department of Computer Science and Automatics, University of Bielsko-Biała, ul. Willowa 2, 43-309 Bielsko-Biała, PolandInstitute of Computational Intelligence, Częstochowa University of Technology, 42-201 Częstochowa, PolandThe purpose of instance selection is to reduce the data size while preserving as much useful information stored in the data as possible and detecting and removing the erroneous and redundant information. In this work, we analyze instance selection in regression tasks and apply the NSGA-II multi-objective evolutionary algorithm to direct the search for the optimal subset of the training dataset and the k-NN algorithm for evaluating the solutions during the selection process. A key advantage of the method is obtaining a pool of solutions situated on the Pareto front, where each of them is the best for certain RMSE-compression balance. We discuss different parameters of the process and their influence on the results and put special efforts to reducing the computational complexity of our approach. The experimental evaluation proves that the proposed method achieves good performance in terms of minimization of prediction error and minimization of dataset size.http://www.mdpi.com/1099-4300/20/10/746instance selectioninformation selectionmulti-objective evolutionary algorithmsregressionk-NNcomputational complexity |
| spellingShingle | Mirosław Kordos Krystian Łapa Multi-Objective Evolutionary Instance Selection for Regression Tasks Entropy instance selection information selection multi-objective evolutionary algorithms regression k-NN computational complexity |
| title | Multi-Objective Evolutionary Instance Selection for Regression Tasks |
| title_full | Multi-Objective Evolutionary Instance Selection for Regression Tasks |
| title_fullStr | Multi-Objective Evolutionary Instance Selection for Regression Tasks |
| title_full_unstemmed | Multi-Objective Evolutionary Instance Selection for Regression Tasks |
| title_short | Multi-Objective Evolutionary Instance Selection for Regression Tasks |
| title_sort | multi objective evolutionary instance selection for regression tasks |
| topic | instance selection information selection multi-objective evolutionary algorithms regression k-NN computational complexity |
| url | http://www.mdpi.com/1099-4300/20/10/746 |
| work_keys_str_mv | AT mirosławkordos multiobjectiveevolutionaryinstanceselectionforregressiontasks AT krystianłapa multiobjectiveevolutionaryinstanceselectionforregressiontasks |