犯罪分布预测问题对社会稳定有积极影响,在学术界引起了广泛关注。现有的研究方法并不能很好的适用于本文所用的数据集。因此,本文构建了矢量运动模型,并提出了一种名为TPML-WMA(Transition Probability Matrix Learning and Weighted M...犯罪分布预测问题对社会稳定有积极影响,在学术界引起了广泛关注。现有的研究方法并不能很好的适用于本文所用的数据集。因此,本文构建了矢量运动模型,并提出了一种名为TPML-WMA(Transition Probability Matrix Learning and Weighted Moving Average)的新算法,利用此算法去学习转移概率矩阵,然后对得到的矩阵做加权移动处理,进而来预测未来的犯罪率分布。本文所用数据集是中国某市2001年到2011年共11年间盗窃案发生的数据集,在此之上本文建立了VM(Vector Motion)模型,提出了TPML-WMA算法对犯罪率进行预测,并讨论了不同初始条件下算法的性能。本文还将所提出的算法与基于最小二乘法的经典线性回归方法进行比较。拟合的结果表明TPML-WMA模型的预测性能较线性回归模型有了很大提升。展开更多
The objective of this study was to investigate and evaluate the rheological and moisture susceptibility of the WMA mixtures containing moist aggregates. The experimental design included the utilizations of two aggrega...The objective of this study was to investigate and evaluate the rheological and moisture susceptibility of the WMA mixtures containing moist aggregates. The experimental design included the utilizations of two aggregate sources, two moisture, two anti-stripping additives, and three WMA additives and control. Several properties of the binders and mixtures were obtained including: viscosity, dynamic shear rheometer, beam bending rheometer of WMA binders, indirect tensile strength (ITS), tensile strength ratio (TSR), and flow values. The experimental results indicate that the WMA additives can result in a slight decrease in viscosity and an increase in failure temperature. In addition, the moisture susceptibility results illustrated that the mixtures containing moist aggregates generally have similar ITS values to other mixtures regardless of HMA or WMA types. Moreover, there were no significant differences in the wet ITS values of WMA mixture amongst four types of mixtures under identical conditions (same moisture and lime contents) except for the mixtures containing liquid anti-stripping additive and Asphamin additive.展开更多
文摘犯罪分布预测问题对社会稳定有积极影响,在学术界引起了广泛关注。现有的研究方法并不能很好的适用于本文所用的数据集。因此,本文构建了矢量运动模型,并提出了一种名为TPML-WMA(Transition Probability Matrix Learning and Weighted Moving Average)的新算法,利用此算法去学习转移概率矩阵,然后对得到的矩阵做加权移动处理,进而来预测未来的犯罪率分布。本文所用数据集是中国某市2001年到2011年共11年间盗窃案发生的数据集,在此之上本文建立了VM(Vector Motion)模型,提出了TPML-WMA算法对犯罪率进行预测,并讨论了不同初始条件下算法的性能。本文还将所提出的算法与基于最小二乘法的经典线性回归方法进行比较。拟合的结果表明TPML-WMA模型的预测性能较线性回归模型有了很大提升。
文摘The objective of this study was to investigate and evaluate the rheological and moisture susceptibility of the WMA mixtures containing moist aggregates. The experimental design included the utilizations of two aggregate sources, two moisture, two anti-stripping additives, and three WMA additives and control. Several properties of the binders and mixtures were obtained including: viscosity, dynamic shear rheometer, beam bending rheometer of WMA binders, indirect tensile strength (ITS), tensile strength ratio (TSR), and flow values. The experimental results indicate that the WMA additives can result in a slight decrease in viscosity and an increase in failure temperature. In addition, the moisture susceptibility results illustrated that the mixtures containing moist aggregates generally have similar ITS values to other mixtures regardless of HMA or WMA types. Moreover, there were no significant differences in the wet ITS values of WMA mixture amongst four types of mixtures under identical conditions (same moisture and lime contents) except for the mixtures containing liquid anti-stripping additive and Asphamin additive.