TRAFFIC FORECASTING EMPLOYING MODELING AND ANALYSIS OF PROBABILISTIC INTERDEPENDENCIES AND CONTEXTUAL DATA

摘要

System and method for description is used for structure forecast model, based on statistical machine learning, it can make the prediction about traffic flows and obstruction, one based on a traffic system is abstract into a set of stochastic variable, including variable, it will solve to have in crucial trouble place and the obstruction of time until until blocking, and represent time quantum. Observational data includes traffic flows and dynamics and other contextual datas, for example, the hour and what day, vacation, school's state, timing and main rally naturally, such as sport events, weather forecast, traffic incident report, and construction and closing report. Forecasting method is for alerting, the figured information of the display about the prediction about the obstruction on desktop on the mobile apparatus, in the route recommendation and in the works of off line and real-time automation.