Machine learning methods for developments of binding kinetic models in predicting protein-ligand dissociation rate constants

Binding kinetic properties of protein–ligand complexes are crucial factors affecting the drug potency.Nevertheless,the current in silico techniques are insufficient in providing accurate and robust predictions for binding kinetic properties.To this end,this work develops a variety of binding kinetic models for predicting a critical binding kinetic property,dissociation rate constant,using eight machine learning(ML)methods(Bayesian Neural Network(BNN),partial least squares regression,Bayesian ridge,Gaussian process regression,principal component regression,random forest,support vector machine,extreme gradient boosting)and the descriptors of the van der Waals/electrostatic interaction energies.These eight models are applied to two case studies involving the HSP90 and RIP1 kinase inhibitors.Both regression results of two case studies indicate that the BNN model has the state-of-the-art prediction accuracy(HSP90:R^(2)_(test)=0:947,MAE_(test)=0.184,rtest=0.976,RMSE_(test)=0.220;RIP1 kinase:R^(2)_(test)=0:745,MAE_(test)=0.188,rtest=0.961,RMSE_(test)=0.290)in comparison with other seven ML models.

Enhancing Secure Development in Globally Distributed Software Product Lines: A Machine Learning-Powered Framework for Cyber-Resilient Ecosystems

Embracing software product lines(SPLs)is pivotal in the dynamic landscape of contemporary software devel-opment.However,the flexibility and global distribution inherent in modern systems pose significant challenges to managing SPL variability,underscoring the critical importance of robust cybersecurity measures.This paper advocates for leveraging machine learning(ML)to address variability management issues and fortify the security of SPL.In the context of the broader special issue theme on innovative cybersecurity approaches,our proposed ML-based framework offers an interdisciplinary perspective,blending insights from computing,social sciences,and business.Specifically,it employs ML for demand analysis,dynamic feature extraction,and enhanced feature selection in distributed settings,contributing to cyber-resilient ecosystems.Our experiments demonstrate the framework’s superiority,emphasizing its potential to boost productivity and security in SPLs.As digital threats evolve,this research catalyzes interdisciplinary collaborations,aligning with the special issue’s goal of breaking down academic barriers to strengthen digital ecosystems against sophisticated attacks while upholding ethics,privacy,and human values.

Prediction of damage potential in mainshock–aftershock sequences using machine learning algorithms

Assessing the potential damage caused by earthquakes is crucial for a community’s emergency response.In this study,four machine learning(ML)methods—random forest,extremely randomized trees,AdaBoost(AB),and gradient boosting(GB)—were employed to develop prediction models for the damage potential of the mainshock(DIMS)and mainshock–aftershock sequences(DIMA).Building structures were modeled using eight single-degree-of-freedom(SDOF)systems with different hysteretic rules.A set of 662 recorded mainshock–aftershock(MS-AS)ground motions was selected from the PEER database.Seven intensity measures(IMs)were chosen to represent the characteristics of the mainshock and aftershock.The results revealed that the selected ML methods can well predict the structural damage potential of the SDOF systems,except for the AB method.The GB model exhibited the best performance,making it the recommended choice for predicting DIMS and DIMA among the four ML models.Additionally,the impact of input variables in the prediction was investigated using the shapley additive explanations(SHAP)method.The high-correlation variables were sensitive to the structural period(T).At T=1.0 s,the mainshock peak ground velocity(PGVM)and aftershock peak ground displacement(PGDA)significantly influenced the prediction of DIMA.When T increased to 5.0 s,the primary high-correlation factor of the mainshock IMs changed from PGVM to the mainshock peak ground displacement(PGDM);however,the highcorrelation variable of the aftershock IMs remained PGDA.The high-correlation factors for DIMS showed trends similar to those of DIMA.Finally,a table summarizing the first and second high-correlation variables for predicting DIMS and DIMA were provided,offering a valuable reference for parameter selection in seismic damage prediction for mainshock–aftershock sequences.

Prognostic prediction models for postoperative patients with stageⅠtoⅢcolorectal cancer based on machine learning

BACKGROUND Colorectal cancer(CRC)is characterized by high heterogeneity,aggressiveness,and high morbidity and mortality rates.With machine learning(ML)algorithms,patient,tumor,and treatment features can be used to develop and validate models for predicting survival.In addition,important variables can be screened and different applications can be provided that could serve as vital references when making clinical decisions and potentially improving patient outcomes in clinical settings.AIM To construct prognostic prediction models and screen important variables for patients with stageⅠtoⅢCRC.METHODS More than 1000 postoperative CRC patients were grouped according to survival time(with cutoff values of 3 years and 5 years)and assigned to training and testing cohorts(7:3).For each 3-category survival time,predictions were made by 4 ML algorithms(all-variable and important variable-only datasets),each of which was validated via 5-fold cross-validation and bootstrap validation.Important variables were screened with multivariable regression methods.Model performance was evaluated and compared before and after variable screening with the area under the curve(AUC).SHapley Additive exPlanations(SHAP)further demonstrated the impact of important variables on model decision-making.Nomograms were constructed for practical model application.RESULTS Our ML models performed well;the model performance before and after important parameter identification was consistent,and variable screening was effective.The highest pre-and postscreening model AUCs 95%confidence intervals in the testing set were 0.87(0.81-0.92)and 0.89(0.84-0.93)for overall survival,0.75(0.69-0.82)and 0.73(0.64-0.81)for disease-free survival,0.95(0.88-1.00)and 0.88(0.75-0.97)for recurrence-free survival,and 0.76(0.47-0.95)and 0.80(0.53-0.94)for distant metastasis-free survival.Repeated cross-validation and bootstrap validation were performed in both the training and testing datasets.The SHAP values of the important variables were consistent with the clinicopathological characteristics of patients with tumors.The nomograms were created.CONCLUSION We constructed a comprehensive,high-accuracy,important variable-based ML architecture for predicting the 3-category survival times.This architecture could serve as a vital reference for managing CRC patients.

Landslide susceptibility mapping using machine learning algorithms and comparison of their performance at Abha Basin,Asir Region,Saudi Arabia

The current study aimed at evaluating the capabilities of seven advanced machine learning techniques(MLTs),including,Support Vector Machine(SVM),Random Forest(RF),Multivariate Adaptive Regression Spline(MARS),Artificial Neural Network(ANN),Quadratic Discriminant Analysis(QDA),Linear Discriminant Analysis(LDA),and Naive Bayes(NB),for landslide susceptibility modeling and comparison of their performances.Coupling machine learning algorithms with spatial data types for landslide susceptibility mapping is a vitally important issue.This study was carried out using GIS and R open source software at Abha Basin,Asir Region,Saudi Arabia.First,a total of 243 landslide locations were identified at Abha Basin to prepare the landslide inventory map using different data sources.All the landslide areas were randomly separated into two groups with a ratio of 70%for training and 30%for validating purposes.Twelve landslide-variables were generated for landslide susceptibility modeling,which include altitude,lithology,distance to faults,normalized difference vegetation index(NDVI),landuse/landcover(LULC),distance to roads,slope angle,distance to streams,profile curvature,plan curvature,slope length(LS),and slope-aspect.The area under curve(AUC-ROC)approach has been applied to evaluate,validate,and compare the MLTs performance.The results indicated that AUC values for seven MLTs range from 89.0%for QDA to 95.1%for RF.Our findings showed that the RF(AUC=95.1%)and LDA(AUC=941.7%)have produced the best performances in comparison to other MLTs.The outcome of this study and the landslide susceptibility maps would be useful for environmental protection.

Gully erosion spatial modelling: Role of machine learning algorithms in selection of the best controlling factors and modelling process

This investigation assessed the efficacy of 10 widely used machine learning algorithms(MLA)comprising the least absolute shrinkage and selection operator(LASSO),generalized linear model(GLM),stepwise generalized linear model(SGLM),elastic net(ENET),partial least square(PLS),ridge regression,support vector machine(SVM),classification and regression trees(CART),bagged CART,and random forest(RF)for gully erosion susceptibility mapping(GESM)in Iran.The location of 462 previously existing gully erosion sites were mapped through widespread field investigations,of which 70%(323)and 30%(139)of observations were arbitrarily divided for algorithm calibration and validation.Twelve controlling factors for gully erosion,namely,soil texture,annual mean rainfall,digital elevation model(DEM),drainage density,slope,lithology,topographic wetness index(TWI),distance from rivers,aspect,distance from roads,plan curvature,and profile curvature were ranked in terms of their importance using each MLA.The MLA were compared using a training dataset for gully erosion and statistical measures such as RMSE(root mean square error),MAE(mean absolute error),and R-squared.Based on the comparisons among MLA,the RF algorithm exhibited the minimum RMSE and MAE and the maximum value of R-squared,and was therefore selected as the best model.The variable importance evaluation using the RF model revealed that distance from rivers had the highest significance in influencing the occurrence of gully erosion whereas plan curvature had the least importance.According to the GESM generated using RF,most of the study area is predicted to have a low(53.72%)or moderate(29.65%)susceptibility to gully erosion,whereas only a small area is identified to have a high(12.56%)or very high(4.07%)susceptibility.The outcome generated by RF model is validated using the ROC(Receiver Operating Characteristics)curve approach,which returned an area under the curve(AUC)of 0.985,proving the excellent forecasting ability of the model.The GESM prepared using the RF algorithm can aid decision-makers in targeting remedial actions for minimizing the damage caused by gully erosion.

Estimation of Potato Biomass and Yield Based on Machine Learning from Hyperspectral Remote Sensing Data

The estimation of potato biomass and yield can optimize the planting pattern and tap the production potential.Based on partial least square(PLSR),multiple linear regression(MLR),support vector machine(SVM),random forest(RF),BP neural network and other machine learning algorithms,the biomass estimation model of potato in different growth stages is constructed by using single variables such as original spectrum,first-order differential spectrum,combined spectrum index and vegetation index(VI)and their coupled combination variables.The accuracy of the models is compared and analyzed,and the best modeling method of biomass in different growth stages is selected.Based on the optimized modeling method,the biomass of each growth stage is estimated,and the yield estimation model of different growth stages is constructed based on the estimation results and the linear regression analysis method,and the accuracy of the model is verified.The results showed that in tuber formation stage,starch accumulation stage and maturity stage,the biomass estimation accuracy based on combination variable was the highest,the best modeling method was MLR and SVM,in tuber growth stage,the best modeling method was MLR,the effect of yield estimation is good.It provides a reference for the algorithm selection of crop biomass and yield models based on machine learning.

A hybrid machine learning approach based study of production forecasting and factors influencing the multiphase flow through surface chokes

Surface chokes are widely utilized equipment installed on wellheads to control hydrocarbon flow rates.Several correlations have been suggested to model the multiphase flow of oil and gas via surface chokes.However,substantial errors have been reported in empirical fitting models and correlations to estimate hydrocarbon flow because of the reservoir's heterogeneity,anisotropism,variance in reservoir fluid characteristics at diverse subsurface depths,which introduces complexity in production data.Therefore,the estimation of daily oil and gas production rates is still challenging for the petroleum industry.Recently,hybrid data-driven techniques have been reported to be effective for estimation problems in various aspects of the petroleum domain.This paper investigates hybrid ensemble data-driven approaches to forecast multiphase flow rates through the surface choke(viz.stacked generalization and voting architectures),followed by an assessment of the impact of input production control variables.Otherwise,machine learning models are also trained and tested individually on the production data of hydrocarbon wells located in North Sea.Feature engineering has been properly applied to select the most suitable contributing control variables for daily production rate forecasting.This study provides a chronological explanation of the data analytics required for the interpretation of production data.The test results reveal the estimation performance of the stacked generalization architecture has outperformed other significant paradigms considered for production forecasting.

Energy characteristics of urban buildings: Assessment by machine learning

Machine learning techniques have attracted more attention as advanced data analytics in building energy analysis.However,most of previous studies are only focused on the prediction capability of machine learning algorithms to provide reliable energy estimation in buildings.Machine learning also has great potentials to identify energy patterns for urban buildings except for model prediction.Therefore,this paper explores energy characteristic of London domestic properties using ten machine learning algorithms from three aspects:tuning process of learning model;variable importance;spatial analysis of model discrepancy.The results indicate that the combination of these three aspects can provide insights on energy patterns for urban buildings.The tuning process of these models indicates that gas use models should have more terms in comparison with electricity in London and the interaction terms should be considered in both gas and electricity models.The rankings of important variables are very different for gas and electricity prediction in London residential buildings,which suggests that gas and electricity use are affected by different physical and social factors.Moreover,the importance levels for these key variables are markedly different for gas and electricity consumption.There are much more important variables for electricity use in comparison with gas use for the importance levels over 40.The areas with larger model discrepancies can be determined using the local spatial analysis based on these machine learning models.These identified areas have significantly different energy patterns for gas and electricity use.More research is required to understand these unusual patterns of energy use in these areas.

Machine Learning Mapping of Soil Apparent Electrical Conductivity on a Research Farm in Mississippi

Open-source and free tools are readily available to the public to process data and assist producers in making management decisions related to agricultural landscapes. On-the-go soil sensors are being used as a proxy to develop digital soil maps because of the data they can collect and their ability to cover a large area quickly. Machine learning, a subcomponent of artificial intelligence, makes predictions from data. Intermixing open-source tools, on-the-go sensor technologies, and machine learning may improve Mississippi soil mapping and crop production. This study aimed to evaluate machine learning for mapping apparent soil electrical conductivity (ECa) collected with an on-the-go sensor system at two sites (i.e., MF2, MF9) on a research farm in Mississippi. Machine learning tools (support vector machine) incorporated in Smart-Map, an open-source application, were used to evaluate the sites and derive the apparent electrical conductivity maps. Autocorrelation of the shallow (ECas) and deep (ECad) readings was statistically significant at both locations (Moran’s I, p 0.001);however, the spatial correlation was greater at MF2. According to the leave-one-out cross-validation results, the best models were developed for ECas versus ECad. Spatial patterns were observed for the ECas and ECad readings in both fields. The patterns observed for the ECad readings were more distinct than the ECas measurements. The research results indicated that machine learning was valuable for deriving apparent electrical conductivity maps in two Mississippi fields. Location and depth played a role in the machine learner’s ability to develop maps.

基于概念性水文模型与长短时记忆模型耦合的汉江上游流域径流模拟

为了探究概念性水文模型(GR4J)与长短时记忆模型(LSTM)耦合过程中物理模型参数反馈机制以及考虑土壤含水量作为中间变量对物理引导机器学习(PIML)模型径流模拟的影响,本研究构建了PIML模型并设置考虑参数反馈、考虑中间变量和两者同时考虑的3种方案(依次简称为H1、H2、H3),以安康站为控制站的汉江上游流域开展实例研究。结果表明:(1)3种PIML模型径流模拟效果均优于LSTM模型,其中验证期纳什系数(NSE)平均提升10.6%,而PIML-H1与PIML-H3模型径流模拟效果优于GR4J模型,验证期NSE平均提升4.2%,其中PIML-H3模型在3种PIML模型中表现最佳,表明基于LSTM和GR4J模型构建的PIML模型对径流模拟效果有所改善,且同时考虑中间变量和参数反馈因素时对径流模拟效果改善最佳;(2)3种PIML模型对低水流量的模拟效果均优于GR4J和LSTM模型,且PIML-H3模型效果最佳,而对于高水流量,3种PIML模型均表现不佳,说明PIML模型往往在模拟低流量事件中更占优势;(3)3种PIML模型的径流模拟效果均表现出较强的季节性变化,PIML-H2和PIML-H3模型相较于PIML-H1模型受到的季节性变化影响更为明显,说明PIML模型模拟径流结果的季节性变化受到中间变量的影响。研究可为汉江上游流域径流模拟预报提供技术支撑。

基于植被指数特征优选的冬小麦叶片含水量估算

为进一步提升利用高光谱数据快速监测叶片含水量的能力,以关中地区冬小麦为研究对象,分别获取2022和2023年孕穗期、抽穗期及灌浆期冬小麦叶片含水量,并同步监测叶片高光谱信息。通过波段组合的形式构建植被指数,并利用相关性分析进行初步筛选,再以ReliefF算法优选得到输入特征变量,然后分别利用随机森林(random forest,RF)、长短期记忆(long short-term memory,LSTM)网络和基于粒子群(particle swarm optimization,PSO)优化的反向传播神经网络(back propagation neural network,BPNN)构建冬小麦叶片含水量估算模型,并进行精度评价。结果表明,通过相关性分析与ReliefF算法结合优选变量,能够较单独通过相关分析明显提升LSTM和PSO-BPNN模型的建模精度,但对RF模型则无法优化变量。相关性分析与ReliefF结合后PSO-BPNN模型在各生育时期均取得最佳预测结果,其中孕穗期、抽穗期和灌浆期验证集r2分别为0.816、0.736和0.806,RMSE分别为0.546%、0.899%和1.531%,NRMSE分别为0.681%、1.195%和2.185%。由此可见,在相关分析的基础上,通过ReliefF算法优选特征变量能够提升特定模型对冬小麦叶片含水量的估算精度,其中对PSO-BPNN模型的效果最好。

Enhancing rectal cancer liver metastasis prediction:Magnetic resonance imaging-based radiomics,bias mitigation,and regulatory considerations

In this article,we comment on the article by Long et al published in the recent issue of the World Journal of Gastrointestinal Oncology.Rectal cancer patients are at risk for developing metachronous liver metastasis(MLM),yet early prediction remains challenging due to variations in tumor heterogeneity and the limitations of traditional diagnostic methods.Therefore,there is an urgent need for noninvasive techniques to improve patient outcomes.Long et al’s study introduces an innovative magnetic resonance imaging(MRI)-based radiomics model that integrates high-throughput imaging data with clinical variables to predict MLM.The study employed a 7:3 split to generate training and validation datasets.The MLM prediction model was constructed using the training set and subsequently validated on the validation set using area under the curve(AUC)and dollar-cost averaging metrics to assess performance,robustness,and generalizability.By employing advanced algorithms,the model provides a non-invasive solution to assess tumor heterogeneity for better metastasis prediction,enabling early intervention and personalized treatment planning.However,variations in MRI parameters,such as differences in scanning resolutions and protocols across facilities,patient heterogeneity(e.g.,age,comorbidities),and external factors like carcinoembryonic antigen levels introduce biases.Additionally,confounding factors such as diagnostic staging methods and patient comorbidities require further validation and adjustment to ensure accuracy and generalizability.With evolving Food and Drug Administration regulations on machine learning models in healthcare,compliance and careful consideration of these regulatory requirements are essential to ensuring safe and effective implementation of this approach in clinical practice.In the future,clinicians may be able to utilize datadriven,patient-centric artificial intelligence(AI)-enhanced imaging tools integrated with clinical data,which would help improve early detection of MLM and optimize personalized treatment strategies.Combining radiomics,genomics,histological data,and demographic information can significantly enhance the accuracy and precision of predictive models.

中国式现代化评价指标体系的构建与应用

科学合理地构建中国式现代化评价指标体系并对其进行实证检验和应用不仅具有紧迫性而且具有必要性,也能为规范研究提供丰富的实践论据。本文通过借鉴已有相关评价指标体系,依据科学性、实用性、全面性、简洁性的指标构建原则,建立了包含经济现代化、政治现代化、文化现代化、社会现代化、生态现代化、军事现代化、人口现代化7个一级指标、29个二级指标、98个三级指标在内的中国式现代化初始评价指标体系。结合机器学习的LASSO筛选变量法,在此基础上再构建出包含7个一级指标、26个二级指标、78个三级指标在内的中国式现代化最终评价指标体系,并根据熵权-TOPSIS法对我国2012~2023年的指标权重和指数进行了分项和综合评价,得出指标权重排序、现代化发展态势、分类重视程度、现代化制约因素等研究结论,继而从统筹推进各领域的协同发展、积极发挥“前十因素”的“火车头”作用、有效补齐迈向高位的制约短板3个方面提出启示对策。

基于可解释机器学习的自适应可变导向车道通行能力提升研究

为了解决交叉口拥堵问题,本研究根据不同时段的交通流量和流向的不均衡性,提升道路通行效率,设计一种基于车道间排队差异的动态左直可变导向车道控制策略(Lane queue differentiation guidance,LQDG),并利用GBDT(Gradient Boosting Decision Tree)模型和SHAP(SHapley Additive exPlanations)建立特征分析框架,解析交叉口延误致因。实验结果显示:①LQDG控制策略与原方案相比行车延误减少73.01%,极大提升交叉口通行效率;②根据GBD T-SHAP分析结果,主线流量对安全时距和停车距离的影响在1200 pcu/h以下呈负相关,大于1200 pcu/h呈正相关。本方法可为缓解交叉口拥堵提供设计方案以及影响通行能力因素的相关性,为优化城市道路通行能力提供一定的理论支撑。

基于特征变量扩展的含气饱和度随机森林预测方法

采用数据驱动的方式,提出了一种基于随机森林机器学习算法训练出含气饱和度地震预测方法,并将该方法应用于中国西部复杂天然气藏中,分别对单井资料和二维地震资料进行了含气饱和度预测与分析。研究结果表明:(1)抽取井旁道纵波速度、横波速度和密度3个弹性参数叠前地震反演结果作为基本特征变量样本,引入边界合成少数类过采样技术对基本特征变量样本和对应的含气饱和度样本进行平衡化处理;利用扩展弹性阻抗结合数学变换自动生成一系列的扩展变量;再利用随机森林对特征变量进行含气饱和度预测重要性排名,并优选重要性较高的特征变量进行含气饱和度随机森林训练。(2)该方法大幅减少了特征变量提取和优选的人工工作量,且有效减少了信息冗余以及因含气饱和度样本不平衡导致的训练偏倚问题,有效增强了随机森林算法在含气饱和度地震预测方面的能力。(3)实际单井应用中预测的含气饱和度与测井解释的含气饱和度的相关系数可达0.9855;在二维地震资料应用中,该方法比基于常规未平衡化的11个弹性参数作为随机森林输入预测出的含气饱和度精度更高。

Q Learning with Quantum Neural Networks

Applying quantum computing techniques to machine learning has attracted widespread attention recently and quantum machine learning has become a hot research topic. There are three major categories of machine learning: supervised, unsupervised, and reinforcement learning (RL). However, quantum RL has made the least progress when compared to the other two areas. In this study, we implement the well-known RL algorithm Q learning with a quantum neural network and evaluate it in the grid world environment. RL is learning through interactions with the environment, with the aim of discovering a strategy to maximize the expected cumulative rewards. Problems in RL bring in unique challenges to the study with their sequential nature of learning, potentially long delayed reward signals, and large or infinite size of state and action spaces. This study extends our previous work on solving the contextual bandit problem using a quantum neural network, where the reward signals are immediate after each action.

Reinforcement Learning with Deep Quantum Neural Networks

The advantage of quantum computers over classical computers fuels the recent trend of developing machine learning algorithms on quantum computers, which can potentially lead to breakthroughs and new learning models in this area. The aim of our study is to explore deep quantum reinforcement learning (RL) on photonic quantum computers, which can process information stored in the quantum states of light. These quantum computers can naturally represent continuous variables, making them an ideal platform to create quantum versions of neural networks. Using quantum photonic circuits, we implement Q learning and actor-critic algorithms with multilayer quantum neural networks and test them in the grid world environment. Our experiments show that 1) these quantum algorithms can solve the RL problem and 2) compared to one layer, using three layer quantum networks improves the learning of both algorithms in terms of rewards collected. In summary, our findings suggest that having more layers in deep quantum RL can enhance the learning outcome.

基于机器学习对呼吸机报警分析

目的 探讨应用机器学习方法对呼吸机在临床使用中的通气类报警进行研究,获得影响报警的重要参数及报警预测模型,识别无效报警并给予临床提示,使临床得以高效应对呼吸机报警,避免造成报警疲劳等消极影响。方法 建立符合标准数据流程的呼吸机数据管理平台,根据单中心的呼吸机报警信息分析特征值,得出重要参数排序;利用超参数调优建模预测报警的真假;用混淆矩阵、受试者工作特征(Receiver Operating Characteristic,ROC)对机器学习模型进行多项指标验证。结果 对测试集5936次通气类报警进行评估,得出无效报警率为88%(召回率为0.88),模型准确度为0.94,精准度为0.78,ROC曲线下面积为0.92,F1得分为0.82。结论 采用机器学习便于临床单中心数据建模,能够及时分析获得呼吸机真实警报的重要参数及报警预测;通过呼吸机数据管理平台可有效提示临床无效报警,从而减少医护人员的压力,提高医疗质量。

基于组合光谱输入量的土壤全氮反演模型

以福州市土壤全氮为研究对象,基于土壤反射率及其13种变换光谱进行模型输入量的筛选和优化,构建20个组合光谱输入量模型对研究区土壤全氮进行反演。结果表明:20个组合光谱输入量模型的预测精度高且稳定,能准确估算和粗略估算全氮的模型约占60.0%和40.0%,较单一光谱模型预测精度分别提高了59.3%和6.4%。以组合光谱作为输入量的土壤全氮反演模型,能够实现不同变换光谱间的优势互补,不仅提升了模型的预测能力,也增强了模型的稳定性,为土壤全氮高光谱预测提供新思路。