基于人工智能的理赔材料处理方法及计算机设备

By using artificial intelligence for grouping and field extraction, the problem of identification errors caused by differences in the format of claims materials was solved, achieving standardization of materials and proactive risk interception, thereby improving the accuracy and efficiency of claims review.

CN122415237APending Publication Date: 2026-07-17SHANGHAI ZHONGAN XINKE INFORMATION TECH SERVICES CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHANGHAI ZHONGAN XINKE INFORMATION TECH SERVICES CO LTD
Filing Date
2026-06-15
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

In insurance claims processing, the sources of claims materials are diverse, their formats vary greatly, and their content is inconsistent, leading to incorrect judgments in the early material identification stage and affecting the accuracy of subsequent reviews.

Method used

By using artificial intelligence-based methods, the characteristics of claim materials are obtained, grouped, and processed independently through processing channels. Key fields are extracted, and target extraction fields are determined through alias fields and scenario information. Format standardization and multi-dimensional verification are then performed to intercept high-risk cases.

Benefits of technology

It improved the accuracy and efficiency of claims processing, reduced the risk of errors in automatic review, and lowered the cost of manual review.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122415237A_ABST
    Figure CN122415237A_ABST
Patent Text Reader

Abstract

本申请公开了一种基于人工智能的理赔材料处理方法及计算机设备,涉及人工智能技术领域,方法包括根据理赔材料的材料特征对理赔材料分组、分配处理通道;在各处理通道中,根据各处理通道对应的关键字段类型,从各组理赔材料中提取关键字段类型对应的字段,得到提取字段;根据提取字段、别名字段以及理赔案件的场景信息,确定目标提取字段,别名字段与提取字段语义匹配且表达形式不同;基于目标提取字段更新理赔材料,在预设维度上对更新后的理赔材料进行检验;响应于检验结果为通过,理赔案件进入理赔审核流程;响应于检验结果为不通过,对理赔案件执行拦截处理,以阻断理赔案件进入理赔审核流程,采用本方法能够提高理赔案件处理的可靠性。
Need to check novelty before this filing date? Find Prior Art