Wrapper for building classification models using Covering Arrays
Wrapper methods are a type of feature selection method that finds a subset of variables to improve the performance of a classifier by removing redundant and irrelevant variables. The use of a wrapper implies that each time a candidate solution is explored, the classifier is evaluated on the quality...
| Autores principales: | , , , , , |
|---|---|
| Formato: | Journal Article |
| Lenguaje: | Inglés |
| Publicado: |
Institute of Electrical and Electronics Engineers
2019
|
| Acceso en línea: | https://hdl.handle.net/10568/103962 |
| _version_ | 1855531021408141312 |
|---|---|
| author | Dorado, Hugo Andres Cobos, Carlos Torres Jimenez, Jose Burra, Dharani Dhar Mendoza, Martha Jiménez, Daniel |
| author_browse | Burra, Dharani Dhar Cobos, Carlos Dorado, Hugo Andres Jiménez, Daniel Mendoza, Martha Torres Jimenez, Jose |
| author_facet | Dorado, Hugo Andres Cobos, Carlos Torres Jimenez, Jose Burra, Dharani Dhar Mendoza, Martha Jiménez, Daniel |
| author_sort | Dorado, Hugo Andres |
| collection | Repository of Agricultural Research Outputs (CGSpace) |
| description | Wrapper methods are a type of feature selection method that finds a subset of variables to improve the performance of a classifier by removing redundant and irrelevant variables. The use of a wrapper implies that each time a candidate solution is explored, the classifier is evaluated on the quality measures selected (e.g. accuracy or precision). Though robust, this iteration across several candidate solutions can become computationally intensive and time-consuming. In this paper we propose a wrapper, that is based on binary Covering Arrays (CAs), and binary Incremental Covering Arrays (ICAs), that have been widely used for experimental design and fault detection in software and hardware testing. The new wrapper was evaluated with six classifiers on seven data sets. The results show that the CAs and ICAs with strength 6 significantly improve the performance and reduces the number of variables required by the classifier. A comparative analysis of the proposed method against wrappers based on other search approaches such as genetic algorithms (GA) and particle swarm optimization (PSO), shows that the proposed method yields results similar to GA, but not to PSO, with differences to PSO, in accuracy, which in the majority of cases is below 0.04. This lack of accuracy, by which the new wrapper fails to match PSO, is offset by the fact that the user does not need to fine tune algorithm parameters, such as velocity ranges, timing, cognitive coefficient, and social coefficient, while it is also much easier to program in parallel. |
| format | Journal Article |
| id | CGSpace103962 |
| institution | CGIAR Consortium |
| language | Inglés |
| publishDate | 2019 |
| publishDateRange | 2019 |
| publishDateSort | 2019 |
| publisher | Institute of Electrical and Electronics Engineers |
| publisherStr | Institute of Electrical and Electronics Engineers |
| record_format | dspace |
| spelling | CGSpace1039622025-03-13T09:44:37Z Wrapper for building classification models using Covering Arrays Dorado, Hugo Andres Cobos, Carlos Torres Jimenez, Jose Burra, Dharani Dhar Mendoza, Martha Jiménez, Daniel Wrapper methods are a type of feature selection method that finds a subset of variables to improve the performance of a classifier by removing redundant and irrelevant variables. The use of a wrapper implies that each time a candidate solution is explored, the classifier is evaluated on the quality measures selected (e.g. accuracy or precision). Though robust, this iteration across several candidate solutions can become computationally intensive and time-consuming. In this paper we propose a wrapper, that is based on binary Covering Arrays (CAs), and binary Incremental Covering Arrays (ICAs), that have been widely used for experimental design and fault detection in software and hardware testing. The new wrapper was evaluated with six classifiers on seven data sets. The results show that the CAs and ICAs with strength 6 significantly improve the performance and reduces the number of variables required by the classifier. A comparative analysis of the proposed method against wrappers based on other search approaches such as genetic algorithms (GA) and particle swarm optimization (PSO), shows that the proposed method yields results similar to GA, but not to PSO, with differences to PSO, in accuracy, which in the majority of cases is below 0.04. This lack of accuracy, by which the new wrapper fails to match PSO, is offset by the fact that the user does not need to fine tune algorithm parameters, such as velocity ranges, timing, cognitive coefficient, and social coefficient, while it is also much easier to program in parallel. 2019 2019-10-04T19:13:04Z 2019-10-04T19:13:04Z Journal Article https://hdl.handle.net/10568/103962 en Open Access Institute of Electrical and Electronics Engineers Dorado, Hugo; Cobos, Carlos; Torres-Jimenez, Jose; Burra, Dharani Dhar; Mendoza, Martha & Jimenez, Daniel. (2019). Wrapper for building classification models using Covering Arrays. IEEE Access. 1-16 p. |
| spellingShingle | Dorado, Hugo Andres Cobos, Carlos Torres Jimenez, Jose Burra, Dharani Dhar Mendoza, Martha Jiménez, Daniel Wrapper for building classification models using Covering Arrays |
| title | Wrapper for building classification models using Covering Arrays |
| title_full | Wrapper for building classification models using Covering Arrays |
| title_fullStr | Wrapper for building classification models using Covering Arrays |
| title_full_unstemmed | Wrapper for building classification models using Covering Arrays |
| title_short | Wrapper for building classification models using Covering Arrays |
| title_sort | wrapper for building classification models using covering arrays |
| url | https://hdl.handle.net/10568/103962 |
| work_keys_str_mv | AT doradohugoandres wrapperforbuildingclassificationmodelsusingcoveringarrays AT coboscarlos wrapperforbuildingclassificationmodelsusingcoveringarrays AT torresjimenezjose wrapperforbuildingclassificationmodelsusingcoveringarrays AT burradharanidhar wrapperforbuildingclassificationmodelsusingcoveringarrays AT mendozamartha wrapperforbuildingclassificationmodelsusingcoveringarrays AT jimenezdaniel wrapperforbuildingclassificationmodelsusingcoveringarrays |