Smart Parking Management Assessment Using Machine Learning Algorithms and IOT
Authors:-Assistant Professor Meenakshi Thalor, Ishwari Abuj
Abstract-With the growing number of vehicles and limited parking infrastructure, parking space man emerged as a major challenge in urban areas. In this paper, an extensive study of machine l models in an IoT-supported space is given, focusing on proposing an ML-based model that available parking space. The study compares the performance of several models Typed as (KNNs), support vector machines (SVMs), random forest (RF), decision tree (DT), logistic and Naive Bayes (NB) regarding to “precision, recall, accuracy, and F1-score performance results obtained after running ML models on the data with 65% and 85% threshold are com meaningful insights about their efficiency of prediction in parking vacancy. Random Forest (RF) model shows the best performance based on those metrics in all evalu high precision, recall, accuracy and F1-score values. The IoT-enabled environment shows t showing its effectiveness in falsely predicting parking space availability. In contrast, K- ne (KNNs), decision tree (DT), logistic regression (LR), predicting Naive Bayes (NB) with co exhibit relatively lower performance in crowded parking GLES scenarios. The paper ends deployment of intelligent predictive models, especially random forest, improves substantial and performance of smart parking system as well as it frees waiting time for cars, and henc parking resource utility as well as it decreases real-time travel congestion and increases use environments.
