Predicting parking occupancy via machine learning in the web of things

• DOI: 10.1016/j.iot.2020.100301
• Publication Date: 2020-09-28T18:09:17Z
• Authors: Jesper C. Provoost; Andreas Kamilaris; Luc J.J. Wismans; Sander J. van der...; Maurice van Keulen
• Keywords: Web of Things; Occupancy

The Web of Things (WoT) enables information gathered by sensors deployed in urban environments to be easily shared utilizing open standards and semantic technologies, creating easier integration with other Web-based information, towards advanced knowledge. Besides WoT, an essential aspect understanding dynamic systems is artificial intelligence (AI). Via AI, data produced WoT-enabled sensory observations can analyzed transformed into meaningful which describes predicts current future situations time space. This paper examines the impact WoT AI smart cities, considering a real-world problem, one predicting parking availability. Traffic cameras are used as sensors, together weather forecasting services. Machine learning (ML) employed for analysis, using predictive models based on neural networks random forests. performance ML prediction occupancy better than state art work problem under study, scoring MSE 7.18 at horizon 60 minutes.