Analysis of Automobile Wheel Counting using Novel adaboosting Algorithm with Accuracy Compared to Logistic Regression Algorithm

Abstract

Aim: In order to determine the accuracy of a realtime traffic management system, this work compares novel adaboosting and logistic regression methods to forecast the AutoMobile Wheel Movement Counting. Materials and Methods: The dataset utilized in this article contains 12 columns or attributes and a total of 10,684 rows. The columns in the dataset are named Car Wheels, Bicycle Wheels, Motorcycle Wheels, and Truck Wheels. The data source link provided a sample size of 1,340 records. A Novel adaboosting algorithm (N=20) and Logistic regression (N=20) iterations are simulated by various parameters and automate vehicle monitoring systems to optimize the pH. The 40 iterations were calculated using CilnCal with G power 80% and CI of 95%. Results: Based on obtained results Novel adaboosting Algorithm has significantly better accuracy (84.71%) compared to Logistic regression Algorithm accuracy (80.60%). Statistical significance difference between Novel adaboosting and Logistic regression algorithm was found to be p=0.013 (Independent Sample T Test p<0.05). Conclusion: Novel adaboosting algorithms provide better results in Finding Road Traffic counting than Logistic regression algorithms.