A target user recommendation method and device for advertisement putting

Abstract

The embodiment of the invention provides a target user recommendation method and device for advertisement putting, and the method comprises the steps of obtaining a plurality of data units in advance,and carrying out the clustering of the plurality of data units according to the industry; obtaining a target industry of advertisement putting, and selecting a set number of data units with high recommendation degrees in the data classes of the corresponding industries as first to-be-recommended data units according to the target industry; selecting the data unit with the highest similarity withthe data unit with the highest recommendation degree in the data classes of the corresponding industries as a second to-be-recommended data unit; and recommending the first to-be-recommended data unitand the second to-be-recommended data unit to an advertisement user. According to the technical scheme of the invention, the recommendation of the data units has industry pertinence through the setting of clustering and recommendation degrees, and through the recommendation of similar data units among the industries, the view field of advertising of an advertiser is continuously expanded, more possibilities are found, and a simple and effective advertising method is provided for the advertiser.

Description

technical field [0001] The invention relates to the field of Internet advertisements, in particular to a target user recommendation method and device for advertisement delivery. Background technique [0002] The task of social software advertising is to deliver the advertisements placed by advertisers to the right users on time and in volume. [0003] In the prior art, the advertisement delivery end generally adopts a feature-oriented approach to find users, for example, setting the orientation of an advertisement as [Location: Beijing]. These users will be served the aforementioned ads. [0004] Another way is to deliver the advertisement through the group package, that is, the advertiser can select some specific groups of people (that is, the group group package), and only deliver the advertisement to these specific groups of people. The crowd package can be some users collected by the advertiser itself, or it can be purchased through a third party, or you can choose to ...

AI Technical Summary

Problems solved by technology

[0005] The disadvantage of using feature-targeted advertising is that it is difficult to meet the requirements of accuracy and the number of users at the same time, because if you want to deliver accurately, you need to set a lot of targeting dimensions that are very demanding. At this time, users who can meet the conditions number will decrease

For example, when an advertiser places a car advertisement, the targeting condition set is [Location: Beijing] in the above example. Obviously, there will be many people who meet this condition, but not all of these people are interested in cars. To improve the accuracy of orientation, set the orientation condition to [location: Beijing, male, 20-40 years old, interest: car peripherals, consumption level: high consumption], the proportion of people interested in cars has increased, but the absolute number has decreased. will plummet

Images