Monitor lizards(Varanidae)inhabit both the mainland and islands of all geological types and have diversified into an exceptionally wide range of body sizes,thus providing an ideal model for examining the role of mainl...Monitor lizards(Varanidae)inhabit both the mainland and islands of all geological types and have diversified into an exceptionally wide range of body sizes,thus providing an ideal model for examining the role of mainland versus island in driving species evolution.Here we use phylogenetic comparative methods to examine whether a link exists between body size driven diversification and body size-frequency distributions in varanid lizards and to test the hypothesis that island lizards differ from mainland species in evolutionary processes,body size,and life-history traits(offspring number and size).We predict that:1)since body size drives rapid diversification in groups,a link exists between body size-driven diversification and body size-frequency distributions;2)because of various environments on island,island species will have higher speciation,extinction,and dispersal rates,compared with mainland species;3)as a response to stronger intraspecific competition,island species will maximize individual ability associated with body size to outcompete closely-related species,and island species will produce smaller clutches of larger eggs to increase offspring quality.Our results confirm that the joint effect of differential macroevolutionary rates shapes the species richness pattern of varanid lizards.There is a link between body size-driven diversification and body sizefrequency distributions,and the speciation rate is maximized at medium body sizes.Island species will have higher speciation,equal extinction,and higher dispersal rates compared with mainland species.Smaller clutch size and larger hatchling in the island than in mainland species indicate that offspring quality is more valuable than offspring quantity for island varanids.展开更多
To solve the high-dimensionality issue and improve its accuracy in credit risk assessment,a high-dimensionality-trait-driven learning paradigm is proposed for feature extraction and classifier selection.The proposed p...To solve the high-dimensionality issue and improve its accuracy in credit risk assessment,a high-dimensionality-trait-driven learning paradigm is proposed for feature extraction and classifier selection.The proposed paradigm consists of three main stages:categorization of high dimensional data,high-dimensionality-trait-driven feature extraction,and high-dimensionality-trait-driven classifier selection.In the first stage,according to the definition of high-dimensionality and the relationship between sample size and feature dimensions,the high-dimensionality traits of credit dataset are further categorized into two types:100<feature dimensions<sample size,and feature dimensions≥sample size.In the second stage,some typical feature extraction methods are tested regarding the two categories of high dimensionality.In the final stage,four types of classifiers are performed to evaluate credit risk considering different high-dimensionality traits.For the purpose of illustration and verification,credit classification experiments are performed on two publicly available credit risk datasets,and the results show that the proposed high-dimensionality-trait-driven learning paradigm for feature extraction and classifier selection is effective in handling high-dimensional credit classification issues and improving credit classification accuracy relative to the benchmark models listed in this study.展开更多
电子废弃物回收规模预测是政府制定循环经济发展规划和有关补贴政策、企业进行资源回收价值评估和产能优化的基础。本文考虑电子废弃物回收规模季度数据的季节性数据特征可能导致传统单模型预测误差偏大、预测结果不稳定等问题,基于“分...电子废弃物回收规模预测是政府制定循环经济发展规划和有关补贴政策、企业进行资源回收价值评估和产能优化的基础。本文考虑电子废弃物回收规模季度数据的季节性数据特征可能导致传统单模型预测误差偏大、预测结果不稳定等问题,基于“分解-集成”的思想提出了季节性数据特征驱动的电子废弃物回收规模预测CH-X12/STL-X框架。首先,基于Canova-Hansen(CH)检验对电子废弃物回收规模时间序列的季节性数据特征进行识别,继而对适于进行季节性分解的时间序列采用X12乘法模型或时间序列季节性分解(Seasonal-trend Decomposition Procedure Based on Loess,STL)模型实现季节性分量提取。然后,采用Holt-Winters模型对获得的季节性分量进行预测,并以支持向量回归模型(Support Vector Regression,SVR)预测分解获得的其他分量。最后,通过对各个分量预测结果的线性求和以得到最终的预测结果。实证结果表明,提出CH-X12/STL-X预测框架能够较好地满足不同季节性数据特征驱动的时间序列预测建模需求,且较传统单模型(Holt-Winters模型、季节性差分自回归滑动平均模型、SVR模型)在预测性能上表现良好且稳定。展开更多
基金This work was supported by grants from the Narural Science Foundation of Zhejiang Province to L-H.L.(LY17C030003)National Natural Science Foundation of China to L.-H,L.(31971414)and XJ(31470471)Finance Science and Technology Project of Hainan Province to Y.D.(ZDYF2018219).
文摘Monitor lizards(Varanidae)inhabit both the mainland and islands of all geological types and have diversified into an exceptionally wide range of body sizes,thus providing an ideal model for examining the role of mainland versus island in driving species evolution.Here we use phylogenetic comparative methods to examine whether a link exists between body size driven diversification and body size-frequency distributions in varanid lizards and to test the hypothesis that island lizards differ from mainland species in evolutionary processes,body size,and life-history traits(offspring number and size).We predict that:1)since body size drives rapid diversification in groups,a link exists between body size-driven diversification and body size-frequency distributions;2)because of various environments on island,island species will have higher speciation,extinction,and dispersal rates,compared with mainland species;3)as a response to stronger intraspecific competition,island species will maximize individual ability associated with body size to outcompete closely-related species,and island species will produce smaller clutches of larger eggs to increase offspring quality.Our results confirm that the joint effect of differential macroevolutionary rates shapes the species richness pattern of varanid lizards.There is a link between body size-driven diversification and body sizefrequency distributions,and the speciation rate is maximized at medium body sizes.Island species will have higher speciation,equal extinction,and higher dispersal rates compared with mainland species.Smaller clutch size and larger hatchling in the island than in mainland species indicate that offspring quality is more valuable than offspring quantity for island varanids.
基金This work is partially supported by grants from the Key Program of National Natural Science Foundation of China(NSFC Nos.71631005 and 71731009)the Major Program of the National Social Science Foundation of China(No.19ZDA103).
文摘To solve the high-dimensionality issue and improve its accuracy in credit risk assessment,a high-dimensionality-trait-driven learning paradigm is proposed for feature extraction and classifier selection.The proposed paradigm consists of three main stages:categorization of high dimensional data,high-dimensionality-trait-driven feature extraction,and high-dimensionality-trait-driven classifier selection.In the first stage,according to the definition of high-dimensionality and the relationship between sample size and feature dimensions,the high-dimensionality traits of credit dataset are further categorized into two types:100<feature dimensions<sample size,and feature dimensions≥sample size.In the second stage,some typical feature extraction methods are tested regarding the two categories of high dimensionality.In the final stage,four types of classifiers are performed to evaluate credit risk considering different high-dimensionality traits.For the purpose of illustration and verification,credit classification experiments are performed on two publicly available credit risk datasets,and the results show that the proposed high-dimensionality-trait-driven learning paradigm for feature extraction and classifier selection is effective in handling high-dimensional credit classification issues and improving credit classification accuracy relative to the benchmark models listed in this study.
文摘电子废弃物回收规模预测是政府制定循环经济发展规划和有关补贴政策、企业进行资源回收价值评估和产能优化的基础。本文考虑电子废弃物回收规模季度数据的季节性数据特征可能导致传统单模型预测误差偏大、预测结果不稳定等问题,基于“分解-集成”的思想提出了季节性数据特征驱动的电子废弃物回收规模预测CH-X12/STL-X框架。首先,基于Canova-Hansen(CH)检验对电子废弃物回收规模时间序列的季节性数据特征进行识别,继而对适于进行季节性分解的时间序列采用X12乘法模型或时间序列季节性分解(Seasonal-trend Decomposition Procedure Based on Loess,STL)模型实现季节性分量提取。然后,采用Holt-Winters模型对获得的季节性分量进行预测,并以支持向量回归模型(Support Vector Regression,SVR)预测分解获得的其他分量。最后,通过对各个分量预测结果的线性求和以得到最终的预测结果。实证结果表明,提出CH-X12/STL-X预测框架能够较好地满足不同季节性数据特征驱动的时间序列预测建模需求,且较传统单模型(Holt-Winters模型、季节性差分自回归滑动平均模型、SVR模型)在预测性能上表现良好且稳定。
