基于云映射的多粒度语义决策属性识别
发布时间:2018-10-10 15:59
【摘要】:异类多传感器多属性目标识别中的描述性语义和决策属性信息无法直接进行识别判定,为此基于粒层转化的思想,提出了一种语义和决策属性识别方法。首先构造云映射函数,将语义和决策属性粒层统一到区间粒层,再根据灰色关联的思想计算区间化后的语义和决策属性信息与数据库之间的区间关联度,实现了不等粒层属性的粒度计算,最后采用证据推理进行了识别判定。结合语义和决策属性识别算例及对比分析,验证了所提出方法在异类数据模糊转化处理和识别上的有效性。
[Abstract]:The descriptive semantics and decision attribute information in heterogeneous multi-sensor multi-attribute target recognition cannot be recognized directly. Based on the idea of grain-layer transformation, a method of semantic and decision attribute recognition is proposed. Firstly, the cloud mapping function is constructed to unify the semantic and decision attribute granular layers to the interval grained layer, and then the interval correlation degree between the semantic and decision attribute information and the database is calculated according to the grey relation thought. The granularity calculation of unequal grained layer attribute is realized, and finally, the identification and decision are carried out by evidence reasoning. Combined with semantic and decision attribute recognition examples and comparative analysis, the effectiveness of the proposed method in fuzzy transformation and recognition of heterogeneous data is verified.
【作者单位】: 海军航空工程学院电子信息工程系;
【基金】:国家自然科学基金(61032001) 新世纪优秀人才支持计划(NCET-11-0872)~~
【分类号】:TP391.1;C934
本文编号:2262387
[Abstract]:The descriptive semantics and decision attribute information in heterogeneous multi-sensor multi-attribute target recognition cannot be recognized directly. Based on the idea of grain-layer transformation, a method of semantic and decision attribute recognition is proposed. Firstly, the cloud mapping function is constructed to unify the semantic and decision attribute granular layers to the interval grained layer, and then the interval correlation degree between the semantic and decision attribute information and the database is calculated according to the grey relation thought. The granularity calculation of unequal grained layer attribute is realized, and finally, the identification and decision are carried out by evidence reasoning. Combined with semantic and decision attribute recognition examples and comparative analysis, the effectiveness of the proposed method in fuzzy transformation and recognition of heterogeneous data is verified.
【作者单位】: 海军航空工程学院电子信息工程系;
【基金】:国家自然科学基金(61032001) 新世纪优秀人才支持计划(NCET-11-0872)~~
【分类号】:TP391.1;C934
【相似文献】
相关期刊论文 前10条
1 陶志,许宝栋,汪定伟;基于决策属性支持度的知识约简方法[J];东北大学学报;2002年11期
2 刘鹏惠;陈子春;秦克云;;区间值信息系统的决策属性约简[J];计算机工程与应用;2009年28期
3 朱颢东;钟勇;;一种无决策属性的信息系统的属性约简算法[J];小型微型计算机系统;2010年02期
4 伊卫国;赵波;;基于决策属性的关联规则挖掘[J];大连交通大学学报;2007年01期
5 高维春;谭旭;;决策属性未知下的学生评教粗糙集分析[J];计算机工程与应用;2012年09期
6 宋笑雪;张文修;;基于集值决策属性的集值信息系统[J];计算机工程与应用;2007年17期
7 栾方军;王健;邱海斌;刘天波;;多决策属性的网络协议识别算法[J];小型微型计算机系统;2014年02期
8 孙雪姣;刘惊雷;;CP-nets的定性偏好决策及一致性推理[J];计算机科学;2013年02期
9 徐义;王艳平;窦金培;;基于模糊粗糙逻辑的属性约简与规则提取[J];辽宁工业大学学报(自然科学版);2010年05期
10 田树新;吴晓平;王红霞;;一种基于改进的ROUSTIDA算法的数据补齐方法[J];海军工程大学学报;2011年05期
相关硕士学位论文 前1条
1 刘牛;基于加权的AUC方法评估优化[D];安徽工业大学;2011年
,本文编号:2262387
本文链接:https://www.wllwen.com/guanlilunwen/tongjijuecelunwen/2262387.html