User recognition method and device

Abstract

The invention discloses a user recognition method and device. The method comprises the following steps of: obtaining an internet surging log of a user, and extracting equipment IDs and browsing behavior information thereof from the internet surfing log; aiming at every two undetermined equipment IDs, determining appointed feature values according to the browsing behavior information; and if the appointed feature values satisfy an appointed rule or are classified into appointed types according to a pre-learnt classification model, recognizing that the two undetermined equipment IDs belong to asame user. According to the method and device, the fact that different equipment IDs belong to a same user can be realized, and the advertisement putting ROI can be improved for advertisers.

Description

technical field [0001] Embodiments of the present invention relate to but are not limited to the field of network technology, especially a method and device for user identification. Background technique [0002] With the development of science and technology and the gradual improvement of people's living standards, most Internet users now have multiple Internet devices, such as personal computers, tablets and mobile phones, and their Internet behavior is scattered on various devices. If simply It is believed that there are different people behind different devices, and using this to place advertisements will cause a lot of waste of resources, reduce the ROI (Return On Investment, return on investment) of advertisements, and lose the interests of advertisers. Therefore, for this kind of Advertisers have a strong demand for the multi-screen recognition technology that users can know which device they switch to. [0003] At present, the cross-screen recognition technology main...

AI Technical Summary

Problems solved by technology

[0002] With the development of science and technology and the gradual improvement of people's living standards, most Internet users now have multiple Internet devices, such as personal computers, tablets and mobile phones, and their Internet behavior is scattered on various devices. If simply It is believed that there are different people behind different devices, and using this to place advertisements will cause a lot of waste of resources, reduce the ROI (Return On Investment, return on investment) of advertisements, and lose the interests of advertisers. Therefore, for this kind of Advertisers have a strong demand for the multi-screen recognition technology that users can know which device they switch to.

[0003] At present, the cross-screen recognition technology mainly relies on the strong account system of Internet products or the matching of personal IDs. That is, if a user logs in to the same account with different devices, it means that these devices belong to the same person. Large-scale, cross-platform user coverage, only a few domestic giants can use it, and the use of these data may involve user privacy issues

Images