Dynamic State General Fuzzy Automata Under Semantically Similar Environment: An Application-Based Study on Air Pollution for Smart Cities 4.0

Automata theory establishes a formal structure for modeling computation, yet bridging this theory with practical, ever-changing systems requires enhanced flexibility. This paper presents a dynamic state general fuzzy automata (GFA[Formula: see text]) that enables applications to dynamically adjust their states in response to changing input data. This approach enhances the capabilities of existing general fuzzy automata (GFA) models to handle uncertainty in applications where the state set is not predefined. Instead of defining a new GFA, similar states are integrated into the existing automata. Algorithms are proposed to calculate the similarity between states and accommodate these similar states into the current model. Thus, this work bridges the gap between theoretical and practical applications of fuzzy automata in real-world applications such as resource allocation and air quality prediction. The substantial volume of data presents a significant challenge in these fields. The proposed model addresses this by effectively incorporating similar historical data to enhance air pollution forecasting. A machine learning linear regression model is used for air quality index (AQI) prediction to validate this approach. A comparative analysis between the proposed GFA[Formula: see text] model and the linear regression model highlights the performance differences and demonstrates the effectiveness of each model in accurately predicting AQI values.