Partial least squares(PLS) regression is an important linear regression method that efficiently addresses the multiple correlation problem by combining principal component analysis and multiple regression. In this pap...Partial least squares(PLS) regression is an important linear regression method that efficiently addresses the multiple correlation problem by combining principal component analysis and multiple regression. In this paper, we present a quantum partial least squares(QPLS) regression algorithm. To solve the high time complexity of the PLS regression, we design a quantum eigenvector search method to speed up principal components and regression parameters construction. Meanwhile, we give a density matrix product method to avoid multiple access to quantum random access memory(QRAM)during building residual matrices. The time and space complexities of the QPLS regression are logarithmic in the independent variable dimension n, the dependent variable dimension w, and the number of variables m. This algorithm achieves exponential speed-ups over the PLS regression on n, m, and w. In addition, the QPLS regression inspires us to explore more potential quantum machine learning applications in future works.展开更多
Boreal forests play an important role in global environment systems. Understanding boreal forest ecosystem structure and function requires accurate monitoring and estimating of forest canopy and biomass. We used parti...Boreal forests play an important role in global environment systems. Understanding boreal forest ecosystem structure and function requires accurate monitoring and estimating of forest canopy and biomass. We used partial least square regression （PLSR） models to relate forest parameters, i.e. canopy closure density and above ground tree biomass, to Landsat ETM＋ data. The established models were optimized according to the variable importance for projection （VIP） criterion and the bootstrap method, and their performance was compared using several statistical indices. All variables selected by the VIP criterion passed the bootstrap test （p〈0.05）. The simplified models without insignificant variables （VIP 〈1） performed as well as the full model but with less computation time. The relative root mean square error （RMSE%） was 29% for canopy closure density, and 58% for above ground tree biomass. We conclude that PLSR can be an effective method for estimating canopy closure density and above ground biomass.展开更多
牦牛奶粉的掺假检测和产地识别有助于保障食品安全、维护消费者权益,是促进乳制品市场健康发展的重要举措。传统的DNA检测方法和稳定同位素分析技术的检测周期长,难以满足快速、低成本现场分析的需求。针对以上问题,本研究建立了一种基...牦牛奶粉的掺假检测和产地识别有助于保障食品安全、维护消费者权益,是促进乳制品市场健康发展的重要举措。传统的DNA检测方法和稳定同位素分析技术的检测周期长,难以满足快速、低成本现场分析的需求。针对以上问题,本研究建立了一种基于近红外光谱技术(Near-infrared Spectroscopy,NIRS)快速辨别牦牛奶粉掺假及产地的方法。收集了来自四川、甘肃、云南及青海的9个品牌的牦牛奶粉。在制备掺假样品之前,采用聚合酶链式反应(Polymerase Chain Reaction,PCR)技术和DNA凝胶电泳验证所收集的牦牛奶粉中是否掺杂了牛奶粉。完成验证后,进行掺假样品的制备以及近红外光谱数据的采集。采用K最邻近法(K-Nearest Neighbors,KNN)建立分类模型,偏最小二乘回归(Partial Least Squares Regression,PLSR)建立定量预测模型。通过优化光谱预处理方法和变量筛选方法进一步提升定量预测模型的预测能力。结果表明,KNN对牦牛奶粉掺假检测(纯牛奶粉、纯牦牛奶粉、掺杂着牛奶粉的牦牛奶粉)及产地识别(四川、甘肃、云南、青海)实现了100%的正确分类。掺假定量预测模型的校正集相关系数(R_(c))为0.9975,预测集相关系数(R_(p))为0.9913,预测集均方根误差(Root Mean Square Error of Prediction,RMSEP)为1.9823%,性能偏差比(Ratio of Performance to Deviation,RPD)为7.2522。本方法可快速、准确地预测牦牛奶粉中牛奶粉的掺杂以及牦牛奶粉产地的辨别,为牦牛奶粉的质量控制提供技术支持。展开更多
Pinus densiflora is a pine species native to the Korean peninsula,and seed orchards have supplied mate-rial needed for afforestation in South Korea.Climate vari-ables affecting seed production have not been identified...Pinus densiflora is a pine species native to the Korean peninsula,and seed orchards have supplied mate-rial needed for afforestation in South Korea.Climate vari-ables affecting seed production have not been identified.The purpose of this study was to determine climate variables that influence annual seed production of two seed orchards using multiple linear regression(MLR),elastic net regres-sion(ENR)and partial least square regression(PLSR)mod-els.The PLSR model included 12 climatic variables from 2003 to 2020 and explained 74.3%of the total variation in seed production.It showed better predictive performance(R2=0.662)than the EN(0.516)and the MLR(0.366)mod-els.Among the 12 climatic variables,July temperature two years prior to seed production and July precipitation after one year had the strongest influence on seed production.The time periods indicated by the two variables corresponded to pollen cone initiation and female gametophyte development.The results will be helpful for developing seed collection plans,selecting new orchard sites with favorable climatic conditions,and investigating the relationships between seed production and climatic factors in related pine species.展开更多
为分析配料(硫化物、糖)组成对高温芝麻饼粕蛋白酶解物(High temperature sesame cake protein hydrolysate,HTSPH)制备肉味香精的影响,首先固定木糖与HTSPH含量,考察5种硫化物如半胱氨酸(LCys)、甲硫氨酸(L-Met)、硫胺素(VB1)对美拉德...为分析配料(硫化物、糖)组成对高温芝麻饼粕蛋白酶解物(High temperature sesame cake protein hydrolysate,HTSPH)制备肉味香精的影响,首先固定木糖与HTSPH含量,考察5种硫化物如半胱氨酸(LCys)、甲硫氨酸(L-Met)、硫胺素(VB1)对美拉德反应产物(Maillard reaction products,MRPs)的感官特性、挥发性成分、pH、褐变与糖基化程度的影响,随后对比木糖、核糖、半乳糖等9种糖与L-Cys、HTSPH反应得到MRPs的特性差异,最后,混料试验考察L-Cys、木糖、HTSPH三者组成对MRPs感官特性与挥发性成分影响。结果显示,不同硫化物形成MRPs的挥发性成分组成差异明显,L-Cys的MRPs肉味突出,感官评价得分最高(69.32±1.34)。对比不同糖的MRPs的性质组成发现,戊糖的MRPs中感官评价得分显著高于已糖的MRPs,其挥发性成分中富含肉香味成分,木糖是适宜由HTSPH制作肉味香精的糖类。混料试验表明L-Cys、木糖与HTSPH组成会显著影响MRPs风味特征。偏最小二乘回归分析发现22种关键挥发性成分。5-甲基-2-噻吩甲醛、二糠基二硫醚、5-乙基噻吩-2-甲醛、4,6-二甲基-1H,3H-噻吩并[3,4-c]噻吩、苯甲醛、2-糠基2-甲基-3-呋喃基二硫化物、2-糠硫醇与感官得分正相关。本研究为高温芝麻饼粕制备肉味香精提供理论指导。展开更多
基金Project supported by the Fundamental Research Funds for the Central Universities, China (Grant No. 2019XD-A02)the National Natural Science Foundation of China (Grant Nos. U1636106, 61671087, 61170272, and 92046001)+2 种基金Natural Science Foundation of Beijing Municipality, China (Grant No. 4182006)Technological Special Project of Guizhou Province, China (Grant No. 20183001)the Foundation of Guizhou Provincial Key Laboratory of Public Big Data (Grant Nos. 2018BDKFJJ016 and 2018BDKFJJ018)。
文摘Partial least squares(PLS) regression is an important linear regression method that efficiently addresses the multiple correlation problem by combining principal component analysis and multiple regression. In this paper, we present a quantum partial least squares(QPLS) regression algorithm. To solve the high time complexity of the PLS regression, we design a quantum eigenvector search method to speed up principal components and regression parameters construction. Meanwhile, we give a density matrix product method to avoid multiple access to quantum random access memory(QRAM)during building residual matrices. The time and space complexities of the QPLS regression are logarithmic in the independent variable dimension n, the dependent variable dimension w, and the number of variables m. This algorithm achieves exponential speed-ups over the PLS regression on n, m, and w. In addition, the QPLS regression inspires us to explore more potential quantum machine learning applications in future works.
基金supported by the 948 Program of the State Forestry Administration (2009-4-43)the National Natura Science Foundation of China (No.30870420)
文摘Boreal forests play an important role in global environment systems. Understanding boreal forest ecosystem structure and function requires accurate monitoring and estimating of forest canopy and biomass. We used partial least square regression （PLSR） models to relate forest parameters, i.e. canopy closure density and above ground tree biomass, to Landsat ETM＋ data. The established models were optimized according to the variable importance for projection （VIP） criterion and the bootstrap method, and their performance was compared using several statistical indices. All variables selected by the VIP criterion passed the bootstrap test （p〈0.05）. The simplified models without insignificant variables （VIP 〈1） performed as well as the full model but with less computation time. The relative root mean square error （RMSE%） was 29% for canopy closure density, and 58% for above ground tree biomass. We conclude that PLSR can be an effective method for estimating canopy closure density and above ground biomass.
文摘牦牛奶粉的掺假检测和产地识别有助于保障食品安全、维护消费者权益,是促进乳制品市场健康发展的重要举措。传统的DNA检测方法和稳定同位素分析技术的检测周期长,难以满足快速、低成本现场分析的需求。针对以上问题,本研究建立了一种基于近红外光谱技术(Near-infrared Spectroscopy,NIRS)快速辨别牦牛奶粉掺假及产地的方法。收集了来自四川、甘肃、云南及青海的9个品牌的牦牛奶粉。在制备掺假样品之前,采用聚合酶链式反应(Polymerase Chain Reaction,PCR)技术和DNA凝胶电泳验证所收集的牦牛奶粉中是否掺杂了牛奶粉。完成验证后,进行掺假样品的制备以及近红外光谱数据的采集。采用K最邻近法(K-Nearest Neighbors,KNN)建立分类模型,偏最小二乘回归(Partial Least Squares Regression,PLSR)建立定量预测模型。通过优化光谱预处理方法和变量筛选方法进一步提升定量预测模型的预测能力。结果表明,KNN对牦牛奶粉掺假检测(纯牛奶粉、纯牦牛奶粉、掺杂着牛奶粉的牦牛奶粉)及产地识别(四川、甘肃、云南、青海)实现了100%的正确分类。掺假定量预测模型的校正集相关系数(R_(c))为0.9975,预测集相关系数(R_(p))为0.9913,预测集均方根误差(Root Mean Square Error of Prediction,RMSEP)为1.9823%,性能偏差比(Ratio of Performance to Deviation,RPD)为7.2522。本方法可快速、准确地预测牦牛奶粉中牛奶粉的掺杂以及牦牛奶粉产地的辨别,为牦牛奶粉的质量控制提供技术支持。
基金supported by the National Institute of Forest Science and by the R&D Program for Forest Science Technology(No.2022458B10-2224-0201)of the Korea Forest Service.
文摘Pinus densiflora is a pine species native to the Korean peninsula,and seed orchards have supplied mate-rial needed for afforestation in South Korea.Climate vari-ables affecting seed production have not been identified.The purpose of this study was to determine climate variables that influence annual seed production of two seed orchards using multiple linear regression(MLR),elastic net regres-sion(ENR)and partial least square regression(PLSR)mod-els.The PLSR model included 12 climatic variables from 2003 to 2020 and explained 74.3%of the total variation in seed production.It showed better predictive performance(R2=0.662)than the EN(0.516)and the MLR(0.366)mod-els.Among the 12 climatic variables,July temperature two years prior to seed production and July precipitation after one year had the strongest influence on seed production.The time periods indicated by the two variables corresponded to pollen cone initiation and female gametophyte development.The results will be helpful for developing seed collection plans,selecting new orchard sites with favorable climatic conditions,and investigating the relationships between seed production and climatic factors in related pine species.
文摘为分析配料(硫化物、糖)组成对高温芝麻饼粕蛋白酶解物(High temperature sesame cake protein hydrolysate,HTSPH)制备肉味香精的影响,首先固定木糖与HTSPH含量,考察5种硫化物如半胱氨酸(LCys)、甲硫氨酸(L-Met)、硫胺素(VB1)对美拉德反应产物(Maillard reaction products,MRPs)的感官特性、挥发性成分、pH、褐变与糖基化程度的影响,随后对比木糖、核糖、半乳糖等9种糖与L-Cys、HTSPH反应得到MRPs的特性差异,最后,混料试验考察L-Cys、木糖、HTSPH三者组成对MRPs感官特性与挥发性成分影响。结果显示,不同硫化物形成MRPs的挥发性成分组成差异明显,L-Cys的MRPs肉味突出,感官评价得分最高(69.32±1.34)。对比不同糖的MRPs的性质组成发现,戊糖的MRPs中感官评价得分显著高于已糖的MRPs,其挥发性成分中富含肉香味成分,木糖是适宜由HTSPH制作肉味香精的糖类。混料试验表明L-Cys、木糖与HTSPH组成会显著影响MRPs风味特征。偏最小二乘回归分析发现22种关键挥发性成分。5-甲基-2-噻吩甲醛、二糠基二硫醚、5-乙基噻吩-2-甲醛、4,6-二甲基-1H,3H-噻吩并[3,4-c]噻吩、苯甲醛、2-糠基2-甲基-3-呋喃基二硫化物、2-糠硫醇与感官得分正相关。本研究为高温芝麻饼粕制备肉味香精提供理论指导。
