为了提升文本聚类效果,改善传统聚类算法在参数设定,稳定性等方面存在的不足,提出新的文本聚类算法TCBIBK(a Text Clustering algorithm Based on Improved BIRCH and K-nearest neighbor)。该算法以BIRCH聚类算法为原型,聚类过程中除...为了提升文本聚类效果,改善传统聚类算法在参数设定,稳定性等方面存在的不足,提出新的文本聚类算法TCBIBK(a Text Clustering algorithm Based on Improved BIRCH and K-nearest neighbor)。该算法以BIRCH聚类算法为原型,聚类过程中除判断文本对象与簇的距离外,增加判断簇与簇之间的距离,采取主动的簇合并或分裂,设置动态的阈值。同时结合KNN分类算法,在保证良好聚类效率前提下提升聚类稳定性,将TCBIBK算法应用于文本聚类,能够提高文本聚类效果。对比实验结果表明,该算法聚类有效性与稳定性都得到较大提高。展开更多
为了解决甘肃省河西区高速公路建设过程中环境监测困难的问题,通过融入大数据技术设计高速公路环境监测系统。采用最大功率点跟踪监测技术中的小控制单元和功率换算的方式设计定点跟踪结构;利用码分多址传输模式将监测设备与大数据库相...为了解决甘肃省河西区高速公路建设过程中环境监测困难的问题,通过融入大数据技术设计高速公路环境监测系统。采用最大功率点跟踪监测技术中的小控制单元和功率换算的方式设计定点跟踪结构;利用码分多址传输模式将监测设备与大数据库相互联系,缩短数据传输时间;对传统谱聚类算法进行改进;应用VMWare Player 16软件模拟监测过程,提高了监测能力。试验结果表明,设计所采用的公路监测系统的监测范围最大为86.7 m2,预警时间为30.16 s,系统结果准确率为97.1%。展开更多
This study is based on the Tong sheep obtained by the random sampling method of typical colonies in the central area of Baishui County in Shaanxi Province, China. An investigation was undertaken to clarify the gene co...This study is based on the Tong sheep obtained by the random sampling method of typical colonies in the central area of Baishui County in Shaanxi Province, China. An investigation was undertaken to clarify the gene constitution of blood protein and nonprotein types of Tong sheep. Twelve genetic markers were examined by starch-gel electrophoresis and cellulose acetate electrophoresis. Polymorphism in Tong sheep was found at the following 10 loci, transferrin （Tf）, alkaline phosphatase （Alp）, leucine aminopeptidase （Lap）, arylesterase （Ary-Es）, hemoglobin-β （Hb-β）, X-protein （X-p）, carbonic anhydrase （CA）, catalase （Cat）, malate dehydrogenase （MDH）, and lysine （Ly）, whereas, albumin （A1） and postalbumin （Po） loci were monomorphic. Genetic approach degree method and phylogenetic relationship clustering method were used to judge the origin and phylogenetic status of Tong sheep. Results from both methods maintained that Tong sheep belonged to the ＂Mongolia group＂, and Mongolia sheep was the origin of Tong sheep. This was also supported by the history of Tong sheep breeding. Compared to the phylogenetic relationship clustering method, the genetic approach degree method was more reliable for the extraction from East and South of Central Asia, and was more effective in reflecting the breeding course of Tong sheep.展开更多
文摘为了提升文本聚类效果,改善传统聚类算法在参数设定,稳定性等方面存在的不足,提出新的文本聚类算法TCBIBK(a Text Clustering algorithm Based on Improved BIRCH and K-nearest neighbor)。该算法以BIRCH聚类算法为原型,聚类过程中除判断文本对象与簇的距离外,增加判断簇与簇之间的距离,采取主动的簇合并或分裂,设置动态的阈值。同时结合KNN分类算法,在保证良好聚类效率前提下提升聚类稳定性,将TCBIBK算法应用于文本聚类,能够提高文本聚类效果。对比实验结果表明,该算法聚类有效性与稳定性都得到较大提高。
文摘为了解决甘肃省河西区高速公路建设过程中环境监测困难的问题,通过融入大数据技术设计高速公路环境监测系统。采用最大功率点跟踪监测技术中的小控制单元和功率换算的方式设计定点跟踪结构;利用码分多址传输模式将监测设备与大数据库相互联系,缩短数据传输时间;对传统谱聚类算法进行改进;应用VMWare Player 16软件模拟监测过程,提高了监测能力。试验结果表明,设计所采用的公路监测系统的监测范围最大为86.7 m2,预警时间为30.16 s,系统结果准确率为97.1%。
基金the International Cooperation Item of the National Natural Science Foundation of China (No. 30213009, 30310103007, 30410103150)Natural Science Foundation of Jiangsu Province of China (No. BK2007556)+1 种基金Basic Natura Science Foundation for Colleges and Universities Jiangsu Province (No. NK051039) the New Century Talent Project of Yangzhou University in China.
文摘This study is based on the Tong sheep obtained by the random sampling method of typical colonies in the central area of Baishui County in Shaanxi Province, China. An investigation was undertaken to clarify the gene constitution of blood protein and nonprotein types of Tong sheep. Twelve genetic markers were examined by starch-gel electrophoresis and cellulose acetate electrophoresis. Polymorphism in Tong sheep was found at the following 10 loci, transferrin （Tf）, alkaline phosphatase （Alp）, leucine aminopeptidase （Lap）, arylesterase （Ary-Es）, hemoglobin-β （Hb-β）, X-protein （X-p）, carbonic anhydrase （CA）, catalase （Cat）, malate dehydrogenase （MDH）, and lysine （Ly）, whereas, albumin （A1） and postalbumin （Po） loci were monomorphic. Genetic approach degree method and phylogenetic relationship clustering method were used to judge the origin and phylogenetic status of Tong sheep. Results from both methods maintained that Tong sheep belonged to the ＂Mongolia group＂, and Mongolia sheep was the origin of Tong sheep. This was also supported by the history of Tong sheep breeding. Compared to the phylogenetic relationship clustering method, the genetic approach degree method was more reliable for the extraction from East and South of Central Asia, and was more effective in reflecting the breeding course of Tong sheep.
