A standard citation relationship identification system and method based on a gravity model

By using gravity models and federated learning techniques, the citation relationships of standards are quantified, which solves the problems of ambiguity and foreign dependence in the development of traditional standards, and realizes the optimization of an efficient and scientific standard system and independent development, thereby improving the standardization level of the robotics industry.

CN121479716BActive Publication Date: 2026-07-21SHANGHAI INST OF QUALITY & STANDARDIZATION
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SHANGHAI INST OF QUALITY & STANDARDIZATION
Filing Date
2025-10-20
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

Traditional standard development relies on expert experience and lacks quantitative analysis tools, resulting in a vague standard system, difficulty in identifying gaps in the field, high dependence on foreign standards, and time-consuming manual analysis that is prone to missing key reference paths.

Method used

A standard citation relationship identification system based on a gravity model is adopted. By quantifying the citation strength between standards, analyzing the evolution of topics and time series characteristics, and combining federated learning technology, cross-enterprise standard data collaborative analysis is achieved, and a data-driven standard system optimization path is constructed.

Benefits of technology

It significantly improves the efficiency of standard system optimization, reduces the risk of dependence on foreign standards, enhances the scientific and forward-looking nature of standard development, achieves cross-institutional collaboration and data security, and provides efficient visualization interaction and decision support.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121479716B_ABST
    Figure CN121479716B_ABST
Patent Text Reader

Abstract

The application discloses a standard citation relationship identification system and method based on a gravity model, and relates to the technical field of citation relationship; a multi-source data acquisition module collects specified standard metadata by using a self-owned database or captures domestic and foreign standards and processes the scanned copies through OCR; a preprocessing module cleans the metadata and extracts subject headings; a gravity model calculation module calculates the gravity value in combination with the cited frequency, subject distance and the like; a subject feature mining module utilizes BERT and HDBSCAN clustering analysis; a visual interactive module constructs a knowledge graph to support multi-dimensional retrieval; further comprising a time series analysis module for predicting the standard life cycle, a recommendation module for recommending standards based on a collaborative filtering algorithm, and a system dynamic update module for realizing real-time calibration of data; the application improves the optimization efficiency of the standard system, identifies core standards and blanks, reduces dependence on foreign standards, coordinates cross-agency data protection, protects data privacy, improves decision-making efficiency through visualization, and assists in standardization of robots and other fields.
Need to check novelty before this filing date? Find Prior Art