A personalized advertisement pushing method and system based on user portrait

By preprocessing data and analyzing link congestion constraints in the advertising push system, combined with interest-driven candidate compression and revenue-driven collaborative adjudication, the resource competition and latency issues of the advertising push system under high-concurrency requests are resolved, achieving more efficient and stable personalized advertising push.

CN122264865APending Publication Date: 2026-06-23TIANJIN CHUANGXINZHONG TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202610381298.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-26
Publication Date
2026-06-23

AI Technical Summary

Technical Problem

In high-concurrency request scenarios, existing advertising push systems suffer from resource contention and request queuing due to the shared computing resources for candidate ad feature extraction, prediction model invocation, and ranking decisions, which affects system response efficiency and push stability.

Method used

By collecting ad request data, user behavior log data, and push link operation data, and performing preprocessing, push link congestion constraint analysis is conducted. Combined with interest-driven candidate compression and ad revenue collaborative ad decision-making, the size of the candidate ad set is dynamically adjusted and the ranking decision is made to optimize the ad push process.

Benefits of technology

It effectively solves the problem of increased computing resource consumption caused by the expansion of candidate ad scale, improves the matching degree between ad push content and user interests, stabilizes the operation of the push link, and optimizes the benefits and display fairness of ad ranking decisions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122264865A_ABST
    Figure CN122264865A_ABST
Patent Text Reader

Abstract

The application discloses a kind of personalized advertisement pushing method and system based on user portrait, it is related to advertisement pushing technical field.The personalized advertisement pushing method and system based on user portrait, comprising: S1, collection advertisement request data, user behavior log data and push link operation data, obtain candidate advertisement resource data and carry out pre-processing;S2, according to congestion constraint analysis result to carry out load buffer release scheduling control and prediction task concurrent scheduling control;S3, according to candidate compression analysis result to carry out scale compression processing and screening control;S4, according to advertisement income collaborative decision analysis result to carry out sorting and list generation.Solve the existing push method in high concurrency request scene needs to a large number of candidate advertisements to execute feature extraction, prediction model call and sorting decision, since model and sorting service share computing resources, prone to resource competition and request queuing, cause push link delay rise and affect system response efficiency problem.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • A method and system for accurate advertising delivery based on Internet big data

    CN118229362B

  • Intelligent Marketing Method and System Based on User Portrait

    CN118982384B