Designing a prediction model and determining the level of AJA championship sports using Adaptive Neuro-Fuzzy Inference System (ANFIS)

Understanding the position of championship sports in the Armed Forces and the amount of its indicators is very important for managers and hierarchies in this field. Some of the important affairs of the organization, such as the sports budget, decisions and macro-policies of the organization, hiring staff and selecting individuals for some missions based on training, are adopted according to the extent and level of these indicators. The main purpose of this research is to present a software model based on human knowledge of experts and its development by ANFIS. Judgmental sampling method was used to determine the experts, including 21 managers, professors and officials of AJA sports field. AJA championship sports indices were finalized by confirmatory factor analysis. Based on the indicators, AJA sports venues are evaluated by experts from very low to very high. The received data contains a matrix of 210 items. The data matrix is used by considering the modeling default in MATLAB software with Adaptive Neuro-Fuzzy Inference System (ANFIS). To evaluate the performance of the model, the squared parameters of mean square error (RMSE), relative error percentage (ε), mean absolute error (MAE) and coefficient of determination (R2) were used. The values of each are were 0.047, 0.92, 0.066 and 0.889, respectively, which indicate the accuracy and reliability of the model. This research is applied in terms of purpose and survey type according to the data collection method. Based on the model sensitivity test, indicators 7, 8, 9, 10 and 11 are more important and effective depending on the level of the organization.