利用AQL的逐批海洋大数据质量检验模型

A Based-on-AQL Quality Inspection Model for Different Lots of Big Marine Data

海洋数据的质量是数据处理和应用的基础,如何准确高效地评价海洋数据的质量是制约其精确有效应用的关键问题之一.质量检验方案主要涉及3个参数,即批量、样本量和接收数,而现有的质量检验方案大多集中于样本量与接收数之间的关系推导,忽略了数据批量对于质量检验方案的影响.此类方案不适用于批量大小不固定的海洋大数据的质量检验.针对该问题,通过基于接收质量限(acceptance quality limit,AQL)提出了符合超几何分布的海洋大数据优化质量检验模型,建立了批量和样本量之间的联系,平衡了数据生产方和使用方对于数据精确度的需求.最后,通过与传统质量检验模型的比较,验证了其对海洋大数据质量检验的有效性.

Data quality guarantees the marine data processing and application.How to design an optimal quality inspection plan fast and control the marine data quality efficiently grows more and more important for the application of big marine data.A quality inspection plan contains three parameters,which are lot size,sample size and acceptance number.Recent studies mainly focus on developing relationships between sample size and acceptance number,while ignoring the influence of lot size which is not suitable for inspecting big marine data.Hence,hypergeometric distribution and the concept of acceptance quality limit(AQL,for short)are led up to the optimal quality inspection plan for big marine data to solve the problem.By comparing the results of the proposed sampling plan with those calculated in terms of the traditional standard,the reliability of the proposed model is validated.