质量安全网络预警分析框架及模型实现--基于互联网消费者质量风险信息的研究

摘要

In network era,it is a new perspective for quality governance to give earlyGwarning of product safety by using consumer releasing information on the Internet.This paper designs an online earlyGwarning model of product safety in three procedures.First,based on some relative research results,this paper refines a general classification paradigm for online quality information,including information on body inj ury,feeling of insecurity and characters changes.Then,followed by literature analysis and practical application,this paper respectively divides the three categories of quality information into 5 risk levels.At last,evidence theory is employed to fuse different information categories with different risk levels,and details of the whole fusion process are introduced.%运用消费者在互联网上发布的质量信息进行质量安全预警，是网络时代质量治理的新视角。文章研究了基于消费者质量安全网络预警分析框架构建与模型实现的问题。从质量信息分类、风险分级以及警情划分三个方面对质量安全网络预警分析框架进行了设计，提炼出对消费者网络质量信息的一般分类范式，即身体伤害类信息、不安全感类信息和性状改变类信息，并结合网络信息的特性，将这三类信息按风险严重程度分别划分为五个不同的级别，基于风险的等级确定了预警警情的划分。采用证据理论从识别框架构建、可信度评价获取与递归融合算法三个步骤详述了预警模型的构建过程，实现对不同类型、不同风险级别质量信息的融合。这项研究为大数据时代质量的网络治理提供了一种思路与方法。