Information prediction method based on deep learning and related equipment

Abstract

The invention discloses an information prediction method based on deep learning, and the method comprises the steps: obtaining LBS information; inputting the LBS information into a clustering model toobtain a first basic feature; converting the first basic feature to obtain a second basic feature; determining a plurality of target grids from the plurality of POI grids; obtaining TF-IDF features and POI category distinguishing features of a plurality of POI categories of the target grid; mapping the plurality of dotting positions to each target grid, and obtaining dotting features; clusteringthe plurality of dotting positions to obtain a plurality of resident points, mapping the plurality of resident points to a plurality of target grids, and obtaining resident point POI features; fusingthe TF-IDF feature, the POI category distinguishing feature, the dotting feature and the resident point POI feature, and obtaining a position interest point feature; and inputting the second basic features and the position interest point features into a pre-trained model to obtain an information prediction result. The invention also provides related equipment. According to the invention, the prediction accuracy of the LBS information based on the user can be improved.

Description

technical field [0001] The present invention relates to the technical field of deep learning, in particular to an information prediction method and related equipment based on deep learning. Background technique [0002] At present, the user's location-based service (Location Based Services, LBS) information is more and more widely used nowadays. According to the user's LBS information, the user's potential behavior habits, activity trajectory and prediction of the relationship between users can be discovered. Wait. [0003] However, in practice, it is found that in the tag prediction based on the user's LBS information, the user's location is generally combined with the POI (Point of Interest) information closest to the location, so as to construct the user's geographical features, and finally Make predictions. In this way, if there are multiple POIs closest to the user's check-in location, it is impossible to determine which POI the user has checked in at this location. T...

AI Technical Summary

Problems solved by technology

In this way, if there are multiple POIs closest to the user's check-in location, it is impossible to determine which POI the user went to when checking in at this location. Therefore, it is impossible to accurately predict the LBS information based on the user.

Images