检索结果-内蒙古大学图书馆

The use of artificial neural networks in the determination of soil grain composition

STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT 2023年第10期37卷 3797-3805页

作者： Sekula, Klaudia Karlowska-Pik, Joanna Kmiecik, Ewa AGH Univ Krakow Mickiewicza 30 Av PL-30059 Krakow Poland Nicolaus Copernicus Univ Torun Torun Poland

The paper presents the possibility of using data mining tools - artificial neural networks - in prediction of hydrometer reading after 24 h in order to limit the duration of the test to 4 h. The authors analysed a database of 693 granulometric composition analysis results of genetically different soils with the use of radial basis function network (RBF) and multilayer perceptron (MLP). The calculations performed showed that it is possible to use MLP to shorten the test time without affecting the quality of the results. The presented accuracy of the model, in the range of 0.55-0.72, allows one to determine the content of silt and clay fractions with an accuracy of 0.49% for equivalent diameter (dT) and 1.50% for percentage of all particles with a diameter smaller than dT (ZT). The results were better than that achieved using linear re-gression models with all predictors (REG), stepwise regression models (SREG), and classification and regression trees (CRT). Taking into account the uncertainty of hydrometric determinations, the obtained forecast values is lower than this uncertainty, therefore neural networks can be used to predict the results of this type of research.

关键词： Determination of soil grain composition linear regression models Stepwise regression models Classification and regression trees Artificial neural networks Radial basis function network and multilayer perceptron

来源：评论

学校读者我要写书评

暂无评论

Modeling the observed center-surround summation in macaque visual area V1

引用

NEUROCOMPUTING 2005年第1期63卷 499-525页

作者： Hegdé, J Felleman, DJ Salk Inst Biol Studies Vis Ctr Lab La Jolla CA 92037 USA Univ Texas Sch Med Dept Neurobiol & Anat Houston TX 77030 USA

The mechanisms of center-surround summation, the process by which visual cortical neurons integrate the inputs from the classical receptive field and the non-classical surround, are poorly understood. We constructed a set of 32 representative center-surround stimuli using a repertoire of four bar types, and recorded the responses of 83 neurons from visual area V1 in two awake, fixating monkeys to each of the stimuli. We then studied, for each cell individually, the extent to which the observed responses of the cell to center-surround stimuli could be accounted for by a linear regression model of center-surround summation. The model hypothesized that the response of a given cell to a given center-surround stimulus is a weighted linear sum of its responses to the four bar types. This model accurately predicted the observed responses to the center-surround stimuli for about two-thirds of V1 cells (56/83, 67%). The ability of the model to predict the observed responses of the cells was not attributable to overfitting or other modeling artifacts, a lack of surround modulation, or a lack of response modulation across different center-surround stimuli. Furthermore, for many cells, the model was able to predict the cell's responses to novel stimuli, indicating that the model captured the center-surround summation behavior of these cells adequately. Together, our results indicate that this simple bottom-up summation mechanism can account for many important center-surround phenomena in V1, including surround inhibition or facilitation, and selectivity for popout or collinear stimuli. (C) 2004 Elsevier B.V. All rights reserved.

关键词： contextual effects figure-ground segregation linear regression models primary visual cortex surround modulation

来源：评论

学校读者我要写书评

暂无评论

A Logspline Estimation for a linear regression Model with an Interval-Censored Continuous Covariate

引用

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION 2014年第10期43卷 2521-2539页

作者： Yang, Yujiao Xu, Song Song, Qiongxia E China Normal Univ Dept Stat & Actuarial Sci Shanghai 200062 Peoples R China Huainan Normal Univ Dept Math Huainan Peoples R China Univ Texas Dallas Dept Math Sci Richardson TX 75080 USA

The study of a linear regression model with an interval-censored covariate, which was motivated by an acquired immunodeficiency syndrome (AIDS) clinical trial, was first proposed by Gomez et al. They developed a likelihood approach, together with a two-step conditional algorithm, to estimate the regression coefficients in the model. However, their method is inapplicable when the interval-censored covariate is continuous. In this article, we propose a novel and fast method to treat the continuous interval-censored covariate. By using logspline density estimation, we impute the interval-censored covariate with a conditional expectation. Then, the ordinary least-squares method is applied to the linear regression model with the imputed covariate. To assess the performance of the proposed method, we compare our imputation with the midpoint imputation and the semiparametric hierarchical method via simulations. Furthermore, an application to the AIDS clinical trial is presented.

关键词： Interval-censored covariate linear regression models Logspline density Noninformative censoring

来源：评论

学校读者我要写书评

暂无评论

The study of the selection of emission lines and plasma operating conditions for efficient internal standardization in inductively coupled plasma optical emission spectrometry

引用

SPECTROCHIMICA ACTA PART B-ATOMIC SPECTROSCOPY 2004年第2期59卷 231-242页

作者： Kola, H Perämäki, P Oulu Univ Dept Chem Oulu 90014 Finland

In inductively coupled plasma optical emission spectrometry matrix effects and drift are usually caused by two major factors, namely changes in the energy transfer between the plasma and sample, and changes in the efficiency of sample aerosol formation and transport. Mathematical model was developed for the behavior of the emission lines as the function of the sample uptake rate and the forward power. This model was utilized to study efficiency of internal standardization in non-robust, semi-robust and robust operating conditions. It was shown that in non-robust conditions internal standardization is extremely difficult, whereas in semi-robust and robust conditions it can be used efficiently. These results are in good agreement with those obtained by other authors previously. It was also shown that in general case using only single correction coefficient 'perfect' correction is impossible using traditional internal standardization. Model developed can be used to quantitatively evaluate the efficiency of internal standardization to reduce matrix effects and drift. (C) 2004 Elsevier B.V. All rights reserved.

关键词： inductively coupled plasma optical emission spectrometry matrix effects drift internal standardization linear regression models

来源：评论

学校读者我要写书评

暂无评论

STATISTICAL MODELING OF WASTE-WATER QUALITY - THE CASE OF MICROELECTRONICS INDUSTRY

引用

WATER RESEARCH 1995年第11期29卷 2541-2547页

作者： VELINOVA, RR KOUMANOVA, BK SOFIA TECHNOL UNIV DEPT CHEM ENGNBU-1156 SOFIABULGARIA

Using a passive experiment, wastewaters from a micro-electronics industry were subjected to experimental evaluation and models for effluent quality prediction were generated. The parameters total solids (TS105) and concentration of Cr(VI) were used as dependent variables. Taking into account both system and experiment particularities, a detailed statistical procedure for model formulation, calibration and verification has been applied. Although the equations thus obtained are specific for the studied system, the procedure used for their development could be considered as applicable for chemometrics purposes.

关键词： INDUSTRIAL WASTEWATERS CHEMOMETRICS linear regression models TOTAL SOLIDS CR(VI)

来源：评论

学校读者我要写书评

暂无评论

Online Updating of Statistical Inference in the Big Data Setting

引用

TECHNOMETRICS 2016年第3期58卷 393-403页

作者： Schifano, Elizabeth D. Wu, Jing Wang, Chun Yan, Jun Chen, Ming-Hui Univ Connecticut Dept Stat Storrs CT 06269 USA

We present statistical methods for big data arising from online analytical processing, where large amounts of data arrive in streams and require fast analysis without storage/access to the historical data. In particular, we develop iterative estimating algorithms and statistical inferences for linear models and estimating equations that update as new data arrive. These algorithms are computationally efficient, minimally storage-intensive, and allow for possible rank deficiencies in the subset design matrices due to rare-event covariates. Within the linear model setting, the proposed online-updating framework leads to predictive residual tests that can be used to assess the goodness of fit of the hypothesized model. We also propose a new online-updating estimator under the estimating equation setting. Theoretical properties of the goodness-of-fit tests and proposed estimators are examined in detail. In simulation studies and real data applications, our estimator compares favorably with competing approaches under the estimating equation setting. Supplementary materials for this article are available online.

关键词： Data compression Data streams Estimating equations linear regression models

来源：评论

学校读者我要写书评

暂无评论

LIDAR-based estimation of bole biomass for precision management of an Amazonian forest: Comparisons of ground-based and remotely sensed estimates

引用

REMOTE SENSING OF ENVIRONMENT 2016年 187卷 281-293页

作者： Figueiredo, Evandro Orfano Neves d'Oliveira, Marcus Vinicio Braz, Evaldo Munoz Papa, Daniel de Almeida Fearnside, Philip Martin Embrapa Acre Rodovia BR-364Km 14 BR-69900056 Rio Branco Acre Brazil Embrapa Florestas Estr Ribeira Km 111CP 319 BR-83411000 Colombo Parana Brazil INPA Ave Andre Araujo 2936 BR-69067375 Manaus Amazonas Brazil Brazilian Res Network Climate Change RedeClima Sao Paulo Brazil

Based on airborne LIDAR data on canopy morphology and height of Amazon forest trees, we developed allometric models to estimate dry biomass stored in the boles of dominant and co-dominant individuals and compared these results with those from equations based on traditional variables such as diameter at breast height (DBH). The database consisted of 142 trees of interest for logging in a forest under management for timber in Brazil's state of Acre. The trees chosen for study were selected through proportional sampling by diameter class (ranging from 45 to 165 cm DBH) in order to properly represent the dominant and co-dominant tree populations with diameters appropriate for harvest. Subsequent to LIDAR profiling of these trees, they were felled, subjected to a battery of dimensional measurements and sampled for wood-density determination. A set of models was generated, followed by model selection and identity testing in order to compare groups of basic wood density (low, medium and high). The morphometric variables of the crown had high explanatory power for bole biomass independent of whether the allometric equations included DBH. When calculating bole biomass with equations that include basic wood density, the best estimate is obtained using variables for both DBH and crown morphology. To obtain an allometric equation that encompasses species in all three classes of basic density, one should either use only independent variables representing crown dimensions or complement these with variables for basic density (BD) and total height (Ht). The study demonstrates the feasibility of using ground-based measurements to calibrate biomass models that include only LIDAR-based variables, allowing much larger areas to be surveyed with reasonable accuracy. The present study is designed to produce data needed for forest management, but the methods developed here can be adapted to studies aimed at reducing the uncertainty in biomass estimates of whole forests (not just harvestabl

关键词： Airborne LIDAR linear regression models Forest profiling Logging Rainforest Tropical forest

来源：评论

学校读者我要写书评

暂无评论

Data-driven flight time prediction for arrival aircraft within the terminal area

引用

IET INTELLIGENT TRANSPORT SYSTEMS 2022年第2期16卷 263-275页

作者： Zhang, Junfeng Peng, Zihan Yang, Chunwei Wang, Bin Nanjing Univ Aeronaut & Astronaut Coll Civil Aviat 29 Jiangjun Ave Nanjing 210016 Jiangsu Peoples R China CAAC Cent & Southern Reg Air Traff Management Bur Guangzhou Peoples R China

Predicting the arrival aircraft's flight time plays a critical role in effectively optimizing and scheduling spatial-temporal resources in the terminal airspace. This paper focuses on a data-driven method for predicting the arrival flight time. First, based on the existing research, a feature set is constructed from four aspects: initial state, arrival pressure, sequencing pressure, and wind information, which are believed to affect arrival flight time significantly. Second, eight widely used models are developed to predict flight time, including linear regression models, nonlinear regression models, and tree-based ensemble models. Furthermore, the stacking technique is adopted to improve the prediction performance. Finally, take Guangzhou Baiyun International Airport as a study case to verify the proposed method's effectiveness. The results indicate that the arrival pressure (describing the arrival traffic demand) and the sequencing pressure (sketching the arrival traffic distribution) could effectively improve the prediction accuracy. The mean absolute percentage error of the predicted flight time via ATAGA and IGONO can be increased by 1%. Besides, the proposed method of extracting wind data could also improve the prediction performance. The mean absolute error of the predicted flight time via GYA can be reduced by 4.85 s.

关键词： data-driven flight time prediction wind data arrival traffic demand arrival traffic distribution trees (mathematics) tree-based ensemble models nonlinear regression models optimisation feature set sequencing pressure air traffic airports forecasting theory mean absolute percentage error terminal airspace arrival pressure stacking technique scheduling linear regression models regression analysis Guangzhou Baiyun International Airport aircraft arrival flight time prediction spatial-temporal resource scheduling ATAGA aircraft IGONO

来源：评论

学校读者我要写书评

暂无评论

Prediction of lichen diversity in an UNESCO biosphere reserve - correlation of high resolution remote sensing data with field samples

引用

ENVIRONMENTAL MODELING & ASSESSMENT 2007年第4期12卷 315-328页

作者： Waser, Lars T. Kuechler, Meinrad Schwarz, Markus Ivits, Eva Stofer, Silvia Scheidegger, Christoph Swiss Fed Res Inst WSL Dept Landscape Invent CH-8903 Birmensdorf Switzerland Univ Freiburg Dept Remote Sensing & LIS D-70106 Freiburg Germany

The present study focuses on developing models to predict lichen species richness in a UNESCO Biosphere Reserve of the Swiss Pre-Alps following a gradient of land-use intensity combining remote sensing data and regression models. The predictive power of the models and the obtained r ranging from 0.5 for lichens on soil to 0.8 for lichens on trees can be regarded as satisfactory to good, respectively. The study revealed that a combination of airborne and spaceborne remote sensing data produced a variety of ecological meaningful variables.

关键词： biodiversity biosphere reserve hotspot lichens linear regression models CIR orthoimages remote sensing

来源：评论

学校读者我要写书评

暂无评论

Prediction error criterion for selecting variables in a linear regression model

引用

ANNALS OF THE INSTITUTE OF STATISTICAL MATHEMATICS 2011年第2期63卷 387-403页

作者： Fujikoshi, Yasunori Kan, Tamio Takahashi, Shin Sakurai, Tetsuro Chuo Univ Dept Math Grad Sch Sci & Engn Bunkyo Ku Tokyo 1128551 Japan Esumi Co Ltd Nakano Ku Tokyo 1640012 Japan

Several criteria, such as CV, C (p) , AIC, CAIC, and MAIC, are used for selecting variables in linear regression models. It might be noted that C (p) has been proposed as an estimator of the expected standardized prediction error, although the target risk function of CV might be regarded as the expected prediction error R (PE). On the other hand, the target risk function of AIC, CAIC, and MAIC is the expected log-predictive likelihood. In this paper, we propose a prediction error criterion, PE, which is an estimator of the expected prediction error R (PE). Consequently, it is also a competitor of CV. Results of this study show that PE is an unbiased estimator when the true model is contained in the full model. The property is shown without the assumption of normality. In fact, PE is demonstrated as more faithful for its risk function than CV. The prediction error criterion PE is extended to the multivariate case. Furthermore, using simulations, we examine some peculiarities of all these criteria.

关键词： Prediction error criterion linear regression models Selection of variables Risk function Selection criteria

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：