Urban activity prediction method based on deep learning

The invention provides a city activity prediction method based on deep learning, and the method is characterized in that the method comprises the following steps: S1, constructing a deep learning model according to a selection model utility function, constructing a training set according to the existing activity data, training the deep learning model through the training set until a training completion condition is reached, and obtaining a training result; if yes, taking the trained deep learning model as a city activity model; s2, collecting resident activity variable data between the residence place of the resident in the specific activity and each to-be-selected destination; s3, constructing a tensor irrelevant to destination attraction and a tensor relevant to the destination attraction according to the resident activity variable data; and S4, inputting the tensor irrelevant to the destination attraction and the tensor relevant to the destination attraction into the city activity model to obtain a corresp