Artificial intelligence-based personalized government service recommendation method and system

By constructing a knowledge graph of government services and dynamic user profiles, and combining service value assessment and causal inference models, personalized government service recommendations are generated. This enables dynamic causal assessment and progressive planning of users' future states, and solves the shortcomings of path planning in existing technologies.

CN122196268APending Publication Date: 2026-06-12CHINA NAT INST OF STANDARDIZATION
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
CHINA NAT INST OF STANDARDIZATION
Filing Date
2026-03-11
Publication Date
2026-06-12

AI Technical Summary

Technical Problem

Existing technologies cannot perform progressive government service path planning based on causal effects, nor can they quantify and assess the causal effect of handling a certain service on a user's future eligibility status, making it difficult to provide government service path planning that is continuous and guiding.

Method used

By constructing a knowledge graph of government services and dynamic user profiles, and using service value assessment functions and counterfactual causal inference models, personalized processing effect vectors are generated. Combined with multi-objective optimization assessment, a progressive planning chain is output.

Benefits of technology

It enables causal quantitative assessment of dynamic changes in users' future states caused by service processing behavior, and provides a paradigm leap from discrete service lists to progressive growth guidance, solving the shortcomings of strategic and continuous path planning in existing technologies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122196268A_ABST
    Figure CN122196268A_ABST
Patent Text Reader

Abstract

The application discloses an artificial intelligence-based personalized government service recommendation method and system, relates to the technical field of personalized service recommendation, and comprises the following steps: constructing a government service knowledge graph based on entities and relations extracted from multiple source policy texts, and obtaining a user dynamic portrait formed by user characteristics and historical records; matching the feature attributes of the user dynamic portrait based on the application conditions of the government service knowledge graph, generating an initial candidate service set, evaluating the value of each service in the initial candidate service set through a service value evaluation function, and obtaining a value evaluation set; and filtering a target service subset from the value evaluation set, and inputting each service in the target service subset into a counterfactual causal inference model. The application solves the problem that the prior art cannot provide strategic and continuous government path planning, and realizes paradigm leap from discrete service list recommendation to progressive growth guidance.
Need to check novelty before this filing date? Find Prior Art