Method and device for data rasterization and method and device for user behavior analysis

Abstract

The embodiment of the invention discloses a method and a device for data rasterization and a method and a device for user behavior analysis. The method for the data rasterization comprises the following steps of acquiring corresponding user associated data according to user behavior data produced in a target time period and located in a target area; computing a two-dimension number corresponding to the user associated data according to position information in the user associated data; acquiring a corresponding two-dimension index number from a two-dimension index number set by taking the two-dimension number as a query condition; associating a raster mark with the user associated data so as to obtain corresponding raster data. The efficiency of the data rasterization is increased, and user behavior rule or preference analysis based on mass of diversified real-time raster data is realized.

Description

technical field [0001] The present invention relates to the field of data processing, and more specifically relates to a method and device for rasterizing data and analyzing user behavior. Background technique [0002] With the development of mobile communication networks, the Internet, mobile terminals (smart phones, car terminals), etc., the server can obtain massive user location data and user behavior data with the help of other nodes in the network. Among them, the user location data is the location of the user at a certain moment collected by the operator through the base station that provides services to the user. For example, Zhang X was at 10:00:00 on October 23, 2014, and his location was longitude 113.5678, latitude 23.111; user behavior data is the behavior attribute data generated by the user through a certain terminal at a certain time, for example, Zhang X made a phone call at 10:00:00 on October 23, 2014, which lasted for 2 minutes, and Baidu user xyz search...

AI Technical Summary

Problems solved by technology

Assuming that m is a million level, k is 10, n is 100, and g is 1000, you need to traverse 10 12 More than times, obviously the time complexity is too high, and the efficiency is low

[0005] Therefore, the current rasterization method cannot handle user behavior data that does not contain user location information. More importantly, extracting data such as user behavior rules or preferences needs to be based on a large number of diverse real-time raster data. The rasterization method needs to occupy a lot of computing resources and is inefficient. It is not suitable for analyzing user behavior rules or preferences, and it cannot satisfy the current site selection of physical commercial institutions or provide quantitative or qualitative reference for commercial promotion. need

Images