Smart City Mobility Application—Gradient Boosting Trees for Mobility Prediction and Analysis Based on Crowdsourced Data

摘要: Mobility management represents one of the most important parts smart city concept. The way we travel, at what time day, for purposes and with transportation modes, have a pertinent impact on overall quality life in cities. To manage this process, detailed comprehensive information individuals’ behaviour is needed as well effective feedback/communication channels. In article, explore applicability crowdsourced data purpose. We apply gradient boosting trees algorithm to model mobility decision making processes (particularly concerning mode they are likely use). accomplish rely collected from three sources: dedicated smartphone application, geographic systems-based web interface weather forecast over period six months. developed seen potential platform personalized cities communication tool between (to steer users towards more sustainable by additionally weighting preferred suggestions) (who can give feedback acceptability provided suggestions, accepting or rejecting them, providing an additional input learning process).