The number of cities, particularly those with advanced infrastructure, is increasing rapidly. There has been a steady increase in the number of automobiles on the road, which has led to severe congestion and wasted time and money. Increasing the number of roads or lanes available is a costly solution to traffic congestion. The primary objective of this research was to examine the traffic pattern using machine learning technologies, which is the optimal method in such situations. The primary objective was to compare the LSTM and ARIMA algorithms across 15-minute intervals, which is confirmed by calculating the observed error. The data obtained was then normalized and filtered to meet the requirements of this study, and machine learning methods are used to make predictions about traffic volume and average speed. Predictions from regression models can be utilized for decision-making. A prediction is a statement about how a variable will change or stay the same. A decision, on the other hand, is what to do in response to a prediction. The LSTM model has less error from the start of the project, while the ARIMA model performance improves with time or at the latter stage. The percentage error of the LSTM model is about 15% less than that of the ARIMA model, hence it can conclude that the LSTM model will perform better than the ARIMA model.
Full Text Finder