Received: 25 April 2023 | Revised: 13 July 2023 | Accepted: 27 July 2023 | Published online: 3 August 2023 RESEARCH ARTICLE A Task Performance and Fitness Predictive Model Based on Neuro-Fuzzy Modeling Femi Johnson 1, * , Onashoga Adebukola 1 , Oluwafolake Ojo 1 , Adejimi Alaba 1 and Opakunle Victor 1 1 Department of Computer Science, Federal University of Agriculture, Nigeria Abstract: Recruiters’ decisions in the selection of candidates for specific job roles are not only dependent on physical attributes and academic qualifications but also on the fitness of candidates for the specified tasks. In this paper, we propose and develop a simple neuro-fuzzy-based task performance and fitness model for the selection of candidates. This is accomplished by obtaining from Kaggle (an online database) samples of task performance-related data of employees in various firms. Data were preprocessed and divided into 60%, 20%, and 20% for training, validating, and testing the developed neuro-fuzzy-based task performance model, respectively. The most significant factors influencing the performance and fitness rating of workers were selected from the database using the principal component analysis (PCA) ranking technique. The effectiveness of the proposed model was assessed and discovered to generate an accuracy of 0.997%, 0.08% root mean square error, and 0.042% mean absolute error. Keywords: neuro-fuzzy, modeling, task performance and fitness performance, prediction, artificial intelligence, practice 1. Introduction Our world is transforming as a result of recent technological breakthroughs, particularly in the manners and ways that tasks are being carried out (Yu et al., 2019). This is also readily apparent in the vast majority of homes and businesses throughout the world where machines and intelligent robots are being used to assist in carrying out specific duties (Muro et al., 2019). In order to save time and produce better results, human energy is purposely directed toward controlling robots and machines, substantially reducing the amount of manual labor that could have been utilized. These have yielded positive outcomes (Tian et al., 2018), which do not only depend on one’s physical attributes and academic standing but also on one’s fitness for the task (Acemoglu & Restrepo, 2019). Recruiters frequently run background checks and take into account a variety of factors when choosing candidates for certain job opportunities (Manasa & Showry, 2018). In most cases, the outcomes of these factors cause several obstacles and issues, including gender segregation among numerous recruiters and the prospective employment roles for the gender of employees (male and female). Late attempts to address the issues could have an impact on the entire organization (Mohr et al., 2017). The examination of pertinent aspects, such as movement restrictions, age differences between spouses, educational gaps, and power influencing task performance, is aided by the use of artificial intelligence (AI) enabled devices. By using AI models, uncertainties and ambiguities caused by incorrect forecasts (Alonso et al., 2018; Graham et al., 2019; Ribeiro et al., 2016) and erroneous ratings of each individual’s task performance rate are removed (Ghafoor et al., 2015). 2. Literature Review Several studies (Kazmi et al., 2017; Sano et al., 2018; Zhao et al., 2019) have expressed differing views on the effectiveness of people’s effort and performance in work environments. A study from the Institute for Women’s Policy Research in 2019 revealed that empowerment measures (including psychological, social, and political freedom as well as autonomy in decision-making) help women perform better over time. The low occurrence or non- existence of the female gender’s participation in decision-making processes in organizations and the society is sometimes referred to as the illegal denial of rights (Hegewisch et al., 2019). According to Obrenovic et al. (2020), the social well-being and safety of individuals affect both genders’ performance. This was revealed in the findings from their developed empirical statistical job performance model with the integration of psychological elements as predictors of work performance. In addition, a five-point Likert scale questionnaire was used to record responses from about two hundred and eighty-seven (287) employees with different educational qualifications and who had spent at least three months in the organization’s main department (accounting, manufacturing, marketing, and human resources management) in 2018. The purpose of choosing employees with educational backgrounds ranging from high school diplomas to master’s degrees was to provide innovative solutions to managers at various levels to help them create a warm and welcoming workplace where employees can fully devote themselves to their careers and improve task performance. *Corresponding author: Femi Johnson, Department of Computer Science, Federal University of Agriculture, Nigeria. Email: femijohnson123@hotmail.com Artificial Intelligence and Applications 2024, Vol. 2(1) 66–72 DOI: 10.47852/bonviewAIA32021010 © The Author(s) 2023. Published by BON VIEW PUBLISHING PTE. LTD. This is an open access article under the CC BY License (https://creativecommons.org/ licenses/by/4.0/). 66