The study focuses on understanding the factors that influence local residents’ support for tourism events, with particular attention given to the differences between Serbia’s two largest urban centers, Belgrade and Novi Sad. The Theory of Planned Behavior (TPB) model was extended by including the construct of community identification in order to capture the affective aspects of social engagement. Data were collected using a combination of online and field surveys, with a total sample of 968 respondents evenly distributed between the two cities. In addition to traditional statistical techniques, machine learning models, Artificial Neural Networks (ANN) and XGBoost, were employed to further validate the structure and significance of the predictors. The additional classification analysis showed that the models were able to successfully differentiate respondents from Belgrade and Novi Sad based on their responses, confirming the existence of structural differences in attitudes and perceptions. The findings suggest that attitude, subjective norms, perceived behavioral control, and community identification are significant predictors of support for tourism events, although their relative importance varies depending on the local context. The research highlights the importance of empowering local communities and fostering a sense of personal involvement as key elements in strengthening the destination’s image and ensuring the sustainable development of tourism events.