Ovalle, ChristianAuccahuasi, WilverMeza, SandraCordova-Buiza, FranklinRojas, KarinCosme, MiryamInciso-Rojas, MiryamAiquipa, GabrielCampos Martínez, Hernando MartinFuentes, AlfonsoAuccahuasi, Aly2023-07-202023-07-202023-01-31Ovalle, C., Auccahuasi, W., Meza, S., Cordova-Buiza, F., Rojas, K., Cosme, M., Inciso-Rojas, M., Aiquipa, G., Campos Martínez, H. M., Fuentes, A., & Auccahuasi, A. (2023). Application of neural networks in the teacher selection process. Procedia Computer Science, 218, 1132-1143. https://doi.org/10.1016/j.procs.2023.01.092https://hdl.handle.net/20.500.12640/3490The information and communications technologies are revolutionizing the classic ways of carrying out the processes, in particular, for the teacher selection processes we have the classic form of evaluation, according to the criteria of each educational institution, in the present work it is presented a teacher selection model, using neural networks, using 3 criteria and 23 characteristics, which are entered into individual networks for each criterion and additionally a network for the final classification, is presented based on a prototype, an application developed with the computational tool Matlab, which is described in detail for its application and scaling, for purposes of measuring the performance of the network, evaluations were carried out with a group of 30 candidates, grouped into two groups, a group of 15 candidates with positive conditions complying with the policies of the educational institution and a second group with candidates who do not meet the policies of the educational institution, with which sensitivity values ​​of 93% and a specificity level of 86% were obtained, we conclude that the model presented can be replicated and conditioned to the needs and policies of each educational institution.application/pdfengAttribution-NonCommercial-NoDerivatives 4.0 Internationalinfo:eu-repo/semantics/openAccessSelectionClassificationNetworkSensitivitySpecificitySelecciónClasificaciónRedSensibilidadEspecificidadinfo:eu-repo/semantics/articlehttps://doi.org/10.1016/j.procs.2023.01.092https://purl.org/pe-repo/ocde/ford#1.02.01