A multi-dimensional knowledge hierarchical modeling and label annotation method for marine documents

By constructing a multi-dimensional knowledge hierarchy model for maritime documents and combining it with employee training data, weaknesses are analyzed, and personalized test question recommendation sequences are generated. This solves the problem of difficulty in identifying employees' weaknesses in traditional methods and realizes dynamic and personalized training for maritime document knowledge modeling and labeling.

CN122412480APending Publication Date: 2026-07-17SHANGHAI COSCO SHIPPING CONTAINER TRANSPORT INFORMATION SERVICE CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHANGHAI COSCO SHIPPING CONTAINER TRANSPORT INFORMATION SERVICE CO LTD
Filing Date
2026-06-16
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Traditional multidimensional knowledge classification modeling and labeling methods for maritime documents are difficult to use in employee training and assessment scenarios to identify specific weaknesses based on the employee's learning process, making it difficult to transform knowledge classification results into a basis for individualized training.

Method used

By collecting multidimensional knowledge source data on maritime documents, a multidimensional knowledge hierarchy model is constructed. Combined with employee training and assessment data, the weaknesses in employees' knowledge system are analyzed, and an intelligent practice question recommendation sequence is set according to the weaknesses to generate personalized test question recommendations.

Benefits of technology

It has achieved integrated processing of maritime document knowledge modeling and labeling, and can dynamically adjust test question recommendations based on employees' learning progress and performance, and conduct personalized training around weak knowledge points, thereby improving the relevance of training.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122412480A_ABST
    Figure CN122412480A_ABST
Patent Text Reader

Abstract

本发明涉及航运单证技术领域,具体为一种面向海运单证的多维知识分级建模与标签标注方法。本发明中,通过先从海运单证业务内容中整理业务流转步骤、任务执行动作和字段定义描述,再将这些内容与培训教材的分级框架建立对应关系,使海运单证知识能够按照业务办理过程逐层组织;随后将员工培训与考核数据引入同一处理链路,根据员工的学习完成情况、历史成绩、绩效评分和考试表现形成学习画像,并将考题文本中的专业术语和业务字段与知识分级模型、目标标签知识库进行匹配。由此,考题标签不再只是对文本本身进行标识,而是能够进一步关联到员工的知识掌握状态。
Need to check novelty before this filing date? Find Prior Art