作者：刘新海, 马彦恒, 侯建强
作者单位：陆军工程大学石家庄校区, 河北 石家庄 050003
关键词：认知测试性;云;频繁项集;数据挖掘;诊断方案
摘要：
针对装备认知测试性智能决策问题，提出基于云和频繁项集的认知测试性诊断方案权衡优化方法。研究装备认知测试性中信息流在定性域和定量域的描述和转换方法，给出基于数据概要的中心云产生方法，实现事务数据清洗与筛选；研究基于频繁项集和新增项集的数据挖掘方法，提出基于2-范数及协方差的数据相关性分析方法，实现基于云和频繁项集的认知测试性诊断方案权衡优化的数据挖掘过程；得到基于存储层-云层-应用层-决策层的认知测试性仿真诊断与权衡优化模型，并对该模型进行补充说明。该方案可为装备认知测试性诊断方案权衡优化的智能化发展奠定基础。
Tradeoff optimization for cognitive testability diagnosis scheme based on cloud and frequent item set
LIU Xinhai, MA Yanheng, HOU Jianqiang
Shijiazhuang Campus of Army Engineering University, Shijiazhuang 050003, China
Abstract: A new kind of tradeoff optimization method for cognitive testability diagnosis scheme based on cloud and frequent item set is proposed to solve intelligent decisions problems in equipment cognitive testability. Focusing on the description and transition method of information flow when researching cognitive testability of equipment in quantitative domain and qualitative domain, a kind of center cloud generation method is presented based on data profile to achieve transaction date cleaning and filtering. The method of data mining based on frequent item set and accessorial item set is also researched. At the same time, an analyzing method using 2-norm and covariance is presented for data correlation analysis, which can achieve the date mining process for tradeoff and optimized diagnosis scheme of cognitive testability based on cloud and frequent item set. A kind of tradeoff and optimized model is gained of database package-cloud package-application package-decision package. At the last, some additional remarks are given besides the model. The research of this paper lays a foundation for intelligent development of tradeoff optimization of cognitive testability diagnosis scheme of equipment.
Keywords: cognitive testability;cloud;frequent item set;data mining;diagnosis scheme
2018, 44(3): 11-15 收稿日期: 2017-08-30;收到修改稿日期: 2017-10-22
基金项目: 武器装备预研基金项目（9140A19031015）
作者简介: 刘新海(1988-),男,山西太原市人,讲师,硕士,研究方向为装备测试性。
参考文献
[1] 马彦恒,刘新海,李刚. 认知测试性的概念及指标分析[J].军械工程学院学报,2016,28(4):62-65.
[2] 王京,李天梅,何华锋,等. 多源测试性综合评估数据等效折合模型与方法研究[J]. 兵工学报,2017,38(1):151-159.
[3] 刘新海. 面向认知电子装备的认知测试性指标体系及关键指标算法研究[D]. 石家庄:军械工程学院,2016.
[4] LEE J, ARDAKANI H D, YANG S, et al. Industrial big data analytics and cyber-physical systems for future maintenance & service innovation[C]//Proceeding of the 4th International Conference on Through-life Engineering Services,2015.
[5] MATHEW S, ALAM M, PECHT M. Identification of failure mechanisms to enhance prognostic outcomes[J]. J Fail Anal and Preven,2012,12(1):66-73.
[6] RECHT M. Nvidia's GPU failure:A case for prognostics and health management[J]. Microelectronics Reliability,2012,52(6):953-957.
[7] LORTON A, FOULADIRAD M, GRALL A. A method-ology for probalistic model-based prognosis[J]. European Journal of Operational Research,2013,225(3):443-454.
[8] QU L D, HE D X, JIA R M. Optimized grey model based on cuckoo search algorithm and its prediction application[J]. Journal of Information & Computational Science,2014(11):1419-1426.
[9] SAKAR C O, KURSUN O. A method for combining mutual information and canonical correlation analysis:predictive mutual information and its use in feature selection[J]. Expert Systems with Applications,2012,39(3):3333-3344.
[10] KURSUN O, ALPAYDIN E, FAVOROV O L, et al. Canonical correlation analysis using within-class coupl-ing[J]. Pattern Recognition Letters,2011,32(2):134-144.
[11] CHAUDHURI K, KAKADE S M, LIVESCU K, et al. Multiview clustering via canonical correlation analysis[C]//Proc of the 26th International Conference on Machine Learning,2009.
[12] 杨静,李文平,张健沛. 基于秩2更新的多维数据流典型相关跟踪算法[J]. 电子学报,2012,40(9):1765-1774.
[13] 杨静,李文平,张健沛. 大数据典型相关分析的云模型方法[J]. 通信学报,2013,34(10):121-134.
[14] 陈槟,尹亚兰,万福. 基于云理论的战术数据链效能评估[J].舰船电子对抗,2015,38(5):64-67.
[15] 彭增焰,吴东,张立敏. 基于Hadoop的频繁项集挖掘算法在图书借阅数据中的应用[J]. 电子技术与软件工程,2017(12):197-198[2017-07-13].http://kns.cnki.net/kcms/detail/10.1108.TP.20170623.1141.300.htm.
[16] 黄超君,范剑波. Apriori算法的分析与改进[J]. 宁波工程学院学报,2013,25(2):64-68.
[17] 毛国君,曹永存. 基于数据概要描述的分布式数据流聚类模型与算法[J]. 计算机科学,2013,40(6):187-191.
