基于关联规则算法的气象微博服务能力多指标评价模型  

A Multi Index Evaluation Model of Meteorological Microblog Service Capability Based on Association Rule Algorithm

作　　者：陈晓宇 吴振鹏[1] 梁家鸿 CHEN Xiaoyu;WU Zhenpeng;LIANG Jiahong(Guangzhou Meteorological Bureau,Guangzhou 511430,China)

机构地区：[1]广州市气象局,广东广州511430

出　　处：《微型电脑应用》2023年第5期137-140,共4页Microcomputer Applications

摘　　要：气象信息具有随机性,导致参与评价气象服务微博服务能力的指标失效,降低了评价结果精度。针对该问题,引入关联规则算法构建气象服务微博服务能力多指标评价模型。采用Apriori算法扫描微博服务的实务数据库,利用关联规则算法挖掘气象微博服务的频繁项,随机选定2个频繁项构建数值区间,控制多指标随机占优度,阻止数值频繁项产生的突变。设定测度数值关系后,统一多指标间的量纲,构建形成多指标评价模型。调用气象服务微博服务后台数据,将其处理为实务数据后,应用不同模型进行对比实验。实验结果表明所设计的多指标评价模型内评价指标产生的失效参数最小,评价模型的评价结果较为准确。The random variability of meteorological information leads to the failure of the indicators participating in the evaluation of meteorological service microblog service capacity,resulting in the reduction of the accuracy of evaluation results.To solve this problem,a multi-index evaluation model of meteorological service microblog service capability is constructed by introducing association rule algorithm.The Apriori algorithm is used to scan the practical database of microblog services,and the association rule algorithm is used to mine the frequent items of microblog services of meteorological services.Two frequent items are randomly selected to build a numerical interval,so as to control the random dominance of multiple indicators and prevent the mutation of numerical frequent items.After setting the measurement value relationship,we unify the dimensions among multiple indicators,and build a multi indicator evaluation model.After calling the background data of meteorological service microblog service and processing it into practical data,different models are applied for comparative experiments.The experimental results show that the failure parameters produced by the evaluation indexes in the designed multi-index evaluation model are the smallest,and the evaluation results of the evaluation model are more accurate.

关 键 词：关联规则 气象微博服务 服务能力 多指标评价模型 

分 类 号：G423.04[文化科学—课程与教学论]