The Effect of Adaptive Neuro-fuzzy Inference System (ANFIS) on Determining the Leadership Perceptions of Construction Employees

In the construction industry, which is Turkey's locomotive and the strategic sector, determining the kind of leadership that impacts employees' productivity is directly related to the success of the business. The identification of leadership types that will motivate and support employees has great importance in terms of construction businesses where the human element is at the forefront. From the point of view of the site chiefs, it is thought that it will benefit all the stakeholders in the construction sector to determine which leader type will motivate which employees. In this study, the productivity relations between the engineers working in construction companies constructing buildings in Adana Province and the employees who are the hierarchically lower-level employees of these persons were investigated using bi-directional surveys. The impact of leadership types on the employees' productivity has been investigated using machine learning. The effects of ANFIS method and the use of genetic algorithm (GA) on the training of ANFIS for the classification are investigated. The data set, which was prepared within the scope of the study, was classified by ANFIS-genetic algorithm (ANFIS-GA), ANFIS-backpropagation algorithm (ANFIS-BP), and ANFIS-hybrid algorithm (ANFIS-HB) algorithms after the required preprocesses. The 10-fold cross-validation technique is used to test the performance of the classification methods. According to the obtained results, the highest accuracy rate of 82.18% is obtained when ANFIS-GA algorithm is used as a classifier. As a result of the study, it is concluded that for this data set, ANFIS, an artificial neural network-based algorithm, is more successful in determining the leadership perceptions of construction employees when it is trained by GA.