OP³ S: On-Street Occupancy based Parking Prices Prediction System for ITS

Now-a-days getting vehicle parked at appropriate location is well known problem in Intelligent Transportation System (ITS). Scarcity of parking space in habitat land pose various type of problems, such as, congestion, pollution, and pricing. In the Seattle city, current issue is congestion which happens due to circling of vehicles in search of parking. Accurate and specific parking information lead to curb congestion and hence pollution. Most of the people are sensitive to prices to be paid for parking their vehicle. Thus, a machine learning based prediction system is proposed in order to predict on street occupancy based parking prices for the Seattle city. In this paper Seattle on-street parking data [1] is considered for training and testing purpose of various machine learning models. In our results, random forest achieved 98.82% accuracy with 0.01 MAE which clearly outperforms other machine learning models.