【摘 要】
：
针对文本相似性度量中的相似度阈值、准确率、召回率、shingle滑动窗口大小、shingle权重系数和文本属性等参数相互影响、关系复杂的问题,研究了这些参数之间的相关性,并结合
论文部分内容阅读
针对文本相似性度量中的相似度阈值、准确率、召回率、shingle滑动窗口大小、shingle权重系数和文本属性等参数相互影响、关系复杂的问题,研究了这些参数之间的相关性,并结合实际应用需求,提出各参数可优化配置的建议,分析与设计了相似度阈值可适应文本篇幅属性的相似性度量算法.通过某基金2009年的7378个项目申请书的比对分析,结果表明:提出的算法不但适用于大规模的文本集合,而且在短小的文本集合中进行相似性度量也具有很高的应用价值,其准确率和召回率均可高达95%以上.
其他文献
Using a fluorescence microscope and EPMA,abundant microbe "bodies" and clear microbic fluorescent microstructure are determined in the ferromanganese nodules re
背景:脐带Whartons Jelly中间充质干细胞可以向胰岛样细胞诱导分化.目的:验证脐带源间充质干细胞与大鼠胰腺细胞共培养向胰岛样细胞诱导分化的可能性,并观察移植后对糖尿病大
The isobaric interference for boron isotopic measurement by negative thermal ionization mass spectrometry (NTIMS) has been studied. The result shows that the CN
A regression model is proposed to relate the variation of water well depth with topographic properties (area and slope), the variation of hydraulic conductivity
In this paper, the Dongchuan-type copper deposits are taken as an example to illustrate the application of GIS to the geo-anomaly-based delineation of mineral r
Natural dynamics can be exploited in the control of biped walking robots: the swing leg can swing passively, the compliant ankle can naturally transfer the cent
报告了12例小儿原发性心脏肿瘩的术后监护.监护要点包括:密切监剥血流动力学变化,保持出入全的平衡,联合应用血苍活性药物,定时监测凝血功能,做好并发症的观察和护理.本组术
On the bases of the field survey and a large number of testing data, we process the qualitative and quantitative assessment of the mine environment in the mid-s
以天津武清地热田为例,对传导型孔隙介质地热田回灌技术进行研究.武清地热田主要开采层为第三系馆陶组砂岩,热储层温度76～86 ℃.根据水动力模型的预测,热储水位降深将在近期趋
The mineral assemblage and content and surface microtextures of slipping zone soil of several landslides in the Three Gorges Reservoir District have been analyz