Accurate predictions, protected privacy
Problem being addressed
Contemporary urban residents, taxi drivers, business sectors, and government agencies have a strong need of accurate and timely traffic flow information as these road users can utilize such information to alleviate traffic congestion, control traffic light appropriately, and improve the efficiency of traffic operations.
Traffic Flow Prediction (TFP) provides traffic flow information by using historical traffic flow data to predict future traffic flow. Nonetheless, it is often overlooked that the data may contain sensitive private information, which leads to potential privacy leakage. The researchers suggest an effective method to accurately predict traffic flow under the constraint of privacy protection. To address the data privacy leakage issue, they incorporate a privacy-preserving machine learning technique for TFP. Such an algorithm provides reliable data privacy preservation through a locally training model without raw data exchange.
Advantages of this solution
Extensive experiments on a real-world dataset demonstrate the performance of the proposed schemes for traffic flow prediction compared to non-federated learning methods. At the moment this is among the pioneering works for traffic flow forecasts with federated deep learning.
Solution originally applied in these industries
Possible New Application of the Work
Travel and Tourism Industry
Accurate traffic prediction can significantly improve the quality of travel planning, especially on a day-to-day basis and in case of large groups of tourists, when optimising the route can save significant time and help avoid traffic congestion.
Source URL: #############