نوع مقاله : مقاله پژوهشی
موضوعات
عنوان مقاله English
نویسندگان English
The development of transportation infrastructure, particularly highways, is one of the fundamental priorities for countries in today's world, and Iran also faces multiple challenges in this area. In this study, the necessary data were initially gathered through text mining of scientific documents. In this process, texts related to highway projects were analyzed from various sources, including scientific articles, reports, and specialized documents, to identify the key factors influencing private sector participation. Based on these factors, a questionnaire was designed and distributed to experts and stakeholders for their input.
After collecting the responses, the data from the questionnaires were analyzed. At this stage, two methods—Structural Equation Modeling (SEM) and the Random Forest algorithm—were employed for data analysis. This study utilizes machine learning methods and Structural Equation Modeling to examine the factors influencing private sector participation in highway projects. SEM, as a powerful tool, analyzes both direct and indirect effects of variables within a complex model using theoretical assumptions and multiple regression equations. On the other hand, the Random Forest algorithm, as an advanced machine learning method, analyzes data and predicts outcomes without requiring strict theoretical assumptions. By creating multiple decision trees and combining their results, this method can identify complex and hidden patterns in the data and predict influential factors with greater accuracy than SEM. The findings indicate that variables related to laws and regulations, as well as financial and tax factors, have the most significant impact on the effectiveness of highway projects. Improving and strengthening these factors can substantially increase the success and efficiency of these projects.
کلیدواژهها English
