分布式多空间数据库复杂时态数据提取技术
|
摘要
传统方法实现过程复杂、历史复杂时态数据的片面性,导致其无法全面地描述时态数据;且相似性计算无法准确匹配具有动态性与复杂性的时态数据,造成提取精度低。为此,提出一种新的分布式多空间数据库复杂时态数据提取技术。设计动态RBF神经网络,对分布式多空间数据库中未知动态进行识别和建模;通过建模结果完成对复杂时态数据的描述。依据加权关联规则与时态关联规则对支持度和置信度的定义,获取T-FS-tree加权时态关联规则中支持度和置信度。将复杂时态数据描述序列、最小支持度、最小置信度作为输入,将加权时态关联规则作为输出,建立T-FS-tree加权时态关联规则挖掘算法。按照向量计算获取加权时态频繁1项集以及频繁2项集,依据获取的加权时态频繁项集建立初始频繁项集树;依据初始频繁项集树获取全部时态频繁项集;通过获取的频繁项集产生加权时态关联规则。从所有关联规则中选择优先度高的规则,构建的复杂时态数据提取器,实现复杂时态数据提取。实验结果表明,所提方法复杂性低,提取结果更加全面、可靠,有很高的准确性。
Traditional methods have complex implementation process,and the one-sidedness of complex historical temporal data makes it impossible to describe temporal data comprehensively,and the similarity calculation can not accurately match the dynamic and complex temporal data,resulting in low extraction accuracy. To solve this problem,a new distributed multi spatial database complex temporal data extraction technology is proposed. Dynamic RBF neural network is designed to identify and model the unknown dynamics in distributed multi-spatial database,and the complex temporal data is described by modeling results. According to the definition of support and confidence of weighted association rules and temporal association rules,the support and confidence of T-FS-tree weighted temporal association rules are obtained. A T-FS-tree weighted temporal association rules mining algorithm is established by taking the complex temporal data description sequence,the minimum support and the minimum confidence as inputs and the weighted temporal association rules as outputs. The weighted temporal frequent itemsets and the frequent itemsets are obtained by vector computation,and the weighted temporal frequent itemsets are obtained by the weighted temporal frequent association rules mining algorithm. Items set up the initial frequent itemset tree,and obtain all the temporal frequent itemsets according to the initial frequent itemset tree,and generate weighted temporal association rules through the obtained frequent itemsets. Selecting high priority rules from all association rules,a complex temporal data extractor is constructed to extract complex temporal data. Experimental results show that the proposed method has low complexity,more comprehensive and reliable extraction results,and high accuracy.
Traditional methods have complex implementation process,and the one-sidedness of complex historical temporal data makes it impossible to describe temporal data comprehensively,and the similarity calculation can not accurately match the dynamic and complex temporal data,resulting in low extraction accuracy. To solve this problem,a new distributed multi spatial database complex temporal data extraction technology is proposed. Dynamic RBF neural network is designed to identify and model the unknown dynamics in distributed multi-spatial database,and the complex temporal data is described by modeling results. According to the definition of support and confidence of weighted association rules and temporal association rules,the support and confidence of T-FS-tree weighted temporal association rules are obtained. A T-FS-tree weighted temporal association rules mining algorithm is established by taking the complex temporal data description sequence,the minimum support and the minimum confidence as inputs and the weighted temporal association rules as outputs. The weighted temporal frequent itemsets and the frequent itemsets are obtained by vector computation,and the weighted temporal frequent itemsets are obtained by the weighted temporal frequent association rules mining algorithm. Items set up the initial frequent itemset tree,and obtain all the temporal frequent itemsets according to the initial frequent itemset tree,and generate weighted temporal association rules through the obtained frequent itemsets. Selecting high priority rules from all association rules,a complex temporal data extractor is constructed to extract complex temporal data. Experimental results show that the proposed method has low complexity,more comprehensive and reliable extraction results,and high accuracy.
引文
1周翔宇,程春玲,杨雁莹.基于分布式内存数据库的移动对象全时态索引[J].计算机科学,2016,43(7):203-207Zhou Xiangyu,Cheng Chunling,Yang Yanying.Full-temporal index of moving objects based on distributed main memory database[J].Computer Science,2016,43(7):203-207
2张玲波,甘元科,石刚,等.同步数据流语言时态消去的可信翻译[J].计算机工程与设计,2014,35(1):137-143Zhang Lingbo,Gan Yuanke,Shi Gang,et al.Certified translation for eliminating temporal feature of synchronous dataflow program[J].Computer Engineering and Design,2014,35(1):137-143
3郜允兵,潘瑜春,高秉博,等.面向土地利用调查的时空数据库构建技术[J].测绘科学,2015,40(5):49-54Gao Yunbing,Pan Yuchun,Gao Bingbo,et al.Key technologies for land use survey oriented spatio-temporal database construction[J].Science of Surveying and Mapping,2015,40(5):49-54
4 Yan L,Roy D P.Automated crop field extraction from multi-temporal web enabled landsat Data[J].Remote Sensing of Environment,2014,144:42-64
5邹保平,黄文思,张文晋,等.基于广义回归神经网络的电网信息系统日志数据分析技术研究[J].电子设计工程,2017,25(13):114-117Zou Baoping,Huang Wensi,Zhang Wenjin,et al.Analysis of data network information system logs based on generalized regression neural network[J].Electronic Design Engineering,2017,25(13):114-117
6陈瑛,叶小平.时态拟序数据索引TQD-tree[J].计算机应用研究,2015,32(3):666-668Chen Ying,Ye Xiaoping.Temporal quasi-order data index TQD-tree[J].Application Research of Computers,2015,32(3):666-668
7曾德伟,吴玉香,王聪.不确定AUV的神经网络辨识和学习控制[J].计算机仿真,2017,34(6):314-318Zeng Dewei,Wu Yuxiang,Wang Cong.Neural network identification and learning control of uncertain AUV[J].Computer Simulation,2017,34(6):314-318
8范开元,米西峰.网络数据包安全指标关联规则挖掘应用与研究[J].科学技术与工程,2014,14(7):216-218Fan Kaiyuan,Mi Xifeng.Network data package communication safety indexes association rules mining research[J].Science Technology and Engineering,2014,14(7):216-218
9陈达伦,陈荣国,谢炯.基于MPP架构的并行空间数据库原型系统的设计与实现[J].地球信息科学学报,2016,18(2):151-159Chen Dalun,Chen Rongguo,Xie Jiong.Research of the parallel spatial database proto system based on MPP architecture[J].Journal of Geo-Information Science,2016,18(2):151-159
10张继荣,王向阳.基于XML数据挖掘的Apriori算法的研究与改进[J].计算机测量与控制,2016,24(6):178-180Zhang Jirong,Wang Xiangyang.Research and improvement of Apriori algorithm for XML data mining[J].Computer Measurement&Control,2016,24(6):178-180
11王育红,张合兵,郭增长.多时态不同差异的土地利用现状数据一致化处理方法[J].中国土地科学,2014,28(12):79-85Wang Yuhong,Zhang Hebing,Guo Zengzhang.Consistency handling approach on multi-temporal land use status data with various differences[J].China Land Science,2014,28(12):79-85
12 Walz U.Extraction of small biotopes and ecotones from multi-temporal RapidEye data and a high-resolution normalized digital surface model[J].International Journal of Remote Sensing,2014,35(20):7245-7262
13李晓东,魏惠茹.支持多模推荐的多层数据库优化访问技术[J].科技通报,2015,31(12):110-112Li Xiaodong,Wei Huiru.Multi database access technology optimization support multimode recommendation[J].Bulletin of Science and Technology,2015,31(12):110-112
14 Cucu-Dumitrescu C,Constantin S.Extraction of regions with similar temporal evolution using earth observation big data application to water turbidity dynamics[J].Remote Sensing Letters,2017,8(7):627-636
15周亮,李格非,邰伟鹏,等.基于Spark的时态查询扩展与时态索引优化研究[J].计算机工程,2017,43(7):22-28Zhou Liang,Li Gefei,Tai Weipeng,et al.Research on temporal query expansion and temporal index optimization based on Spark[J].Computer Engineering,2017,43(7):22-28
2张玲波,甘元科,石刚,等.同步数据流语言时态消去的可信翻译[J].计算机工程与设计,2014,35(1):137-143Zhang Lingbo,Gan Yuanke,Shi Gang,et al.Certified translation for eliminating temporal feature of synchronous dataflow program[J].Computer Engineering and Design,2014,35(1):137-143
3郜允兵,潘瑜春,高秉博,等.面向土地利用调查的时空数据库构建技术[J].测绘科学,2015,40(5):49-54Gao Yunbing,Pan Yuchun,Gao Bingbo,et al.Key technologies for land use survey oriented spatio-temporal database construction[J].Science of Surveying and Mapping,2015,40(5):49-54
4 Yan L,Roy D P.Automated crop field extraction from multi-temporal web enabled landsat Data[J].Remote Sensing of Environment,2014,144:42-64
5邹保平,黄文思,张文晋,等.基于广义回归神经网络的电网信息系统日志数据分析技术研究[J].电子设计工程,2017,25(13):114-117Zou Baoping,Huang Wensi,Zhang Wenjin,et al.Analysis of data network information system logs based on generalized regression neural network[J].Electronic Design Engineering,2017,25(13):114-117
6陈瑛,叶小平.时态拟序数据索引TQD-tree[J].计算机应用研究,2015,32(3):666-668Chen Ying,Ye Xiaoping.Temporal quasi-order data index TQD-tree[J].Application Research of Computers,2015,32(3):666-668
7曾德伟,吴玉香,王聪.不确定AUV的神经网络辨识和学习控制[J].计算机仿真,2017,34(6):314-318Zeng Dewei,Wu Yuxiang,Wang Cong.Neural network identification and learning control of uncertain AUV[J].Computer Simulation,2017,34(6):314-318
8范开元,米西峰.网络数据包安全指标关联规则挖掘应用与研究[J].科学技术与工程,2014,14(7):216-218Fan Kaiyuan,Mi Xifeng.Network data package communication safety indexes association rules mining research[J].Science Technology and Engineering,2014,14(7):216-218
9陈达伦,陈荣国,谢炯.基于MPP架构的并行空间数据库原型系统的设计与实现[J].地球信息科学学报,2016,18(2):151-159Chen Dalun,Chen Rongguo,Xie Jiong.Research of the parallel spatial database proto system based on MPP architecture[J].Journal of Geo-Information Science,2016,18(2):151-159
10张继荣,王向阳.基于XML数据挖掘的Apriori算法的研究与改进[J].计算机测量与控制,2016,24(6):178-180Zhang Jirong,Wang Xiangyang.Research and improvement of Apriori algorithm for XML data mining[J].Computer Measurement&Control,2016,24(6):178-180
11王育红,张合兵,郭增长.多时态不同差异的土地利用现状数据一致化处理方法[J].中国土地科学,2014,28(12):79-85Wang Yuhong,Zhang Hebing,Guo Zengzhang.Consistency handling approach on multi-temporal land use status data with various differences[J].China Land Science,2014,28(12):79-85
12 Walz U.Extraction of small biotopes and ecotones from multi-temporal RapidEye data and a high-resolution normalized digital surface model[J].International Journal of Remote Sensing,2014,35(20):7245-7262
13李晓东,魏惠茹.支持多模推荐的多层数据库优化访问技术[J].科技通报,2015,31(12):110-112Li Xiaodong,Wei Huiru.Multi database access technology optimization support multimode recommendation[J].Bulletin of Science and Technology,2015,31(12):110-112
14 Cucu-Dumitrescu C,Constantin S.Extraction of regions with similar temporal evolution using earth observation big data application to water turbidity dynamics[J].Remote Sensing Letters,2017,8(7):627-636
15周亮,李格非,邰伟鹏,等.基于Spark的时态查询扩展与时态索引优化研究[J].计算机工程,2017,43(7):22-28Zhou Liang,Li Gefei,Tai Weipeng,et al.Research on temporal query expansion and temporal index optimization based on Spark[J].Computer Engineering,2017,43(7):22-28