Advertisement crowd orientation method and device

An advertising and crowd technology, applied in the field of data processing, can solve the problems of high cost, high algorithm complexity, affecting the accuracy of the classifier, etc., to achieve the effect of improving the accuracy and reducing the complexity of the algorithm

Pending Publication Date: 2019-10-18
SHENZHEN TENCENT COMP SYST CO LTD
View PDF0 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The above-mentioned tag user diffusion algorithm, on the one hand, needs to build a classifier for each user tag, the cost of training and maintenance is very high, and the algorithm complexity is high; on the other hand, for some long-tail unpopular user tags, due to the There are fewer original users, which affects the accuracy of the classifier during training

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Advertisement crowd orientation method and device
  • Advertisement crowd orientation method and device
  • Advertisement crowd orientation method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach 1

[0087] If in step 104, users who include each similar user label are used as the targeted advertising groups, when selecting users with the targeted user number from the targeted advertising groups, for K similar user labels, determine the number of users that include each similar user label. User sets, K user sets are obtained, and for each user in the K user sets, the score value of each user is determined according to the user set to which each user belongs and the score value corresponding to each user set, and each user set corresponds to The score value of is the similarity value between the similar user label corresponding to each user set and the target user label, based on the score value of each user, according to the preset rules, select the number of targeted users from the targeted advertising population.

Embodiment approach 2

[0089] If in step 104, the users containing each similar user label and the user containing the target user label are used as the targeted advertising groups, when selecting users with the targeted user number from the targeted advertising groups, for K similar user labels, determine A user set containing each similar user label, to obtain K user sets, for each user in the K user set and each user in the user set containing the target user label, according to the user set to which each user belongs and each user Set the score value corresponding to each user set to determine the score value of each user. The score value corresponding to each user set is the similarity value between the similar user label corresponding to each user set and the target user label, based on the score value of each user, according to The default rule is to select the number of targeted users from the targeted advertising groups.

[0090] Wherein, the preset rule may be a rule with a score value fro...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses an advertisement crowd orientation method and device, belongs to the technical field of data processing. The method and the device are used for considering algorithm complexityand accuracy while realizing label user diffusion during advertisement crowd extraction. The method comprises the following steps: acquiring a target user label which is input by a user and is used for extracting advertisement crowds; respectively determining similarity values between the target user label and other user labels in each pre-acquired user label for describing user characteristics;according to the similarity value between the target user label and other user labels, determining K similar user labels with the maximum similarity with the target user label, and K being a natural number; and taking the user containing each similar user label as the targeted advertisement crowd, or taking the user containing each similar user label and the user containing the target user label as the targeted advertisement crowd.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a method and device for targeting advertising groups. Background technique [0002] Tag group extraction is an important component of the Data Management Platform (DMP). Users give a series of user tags as targeting conditions, and extract advertising groups by associating user portraits, that is, through rich user tags such as interests and behaviors. Group and delineate the required user groups. [0003] Usually, the advertising crowd extraction based on user tags can only extract users that contain user tags, that is, extract as many users as there are users that contain user tags. Perform hashtag user diffusion. [0004] The current mainstream label user diffusion method is: use the original user containing the user label as a positive sample, randomly select other users as negative samples (the negative samples can be obtained using methods such as PU-Learning), an...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q30/02G06K9/62
CPCG06Q30/0269G06F18/214
Inventor 杨春风
Owner SHENZHEN TENCENT COMP SYST CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap