Is the introduction of benthic species necessary for open-water chemical reconstruction in diatom-based transfer functions?

2002 ◽  
Vol 59 (6) ◽  
pp. 938-951 ◽  
Author(s):  
Aline Philibert ◽  
Yves T Prairie
Keyword(s):  
Transfer Functions ◽  
Model Performance ◽  
Open Water ◽  
Partial Least Square ◽  
Least Square ◽  
Weighted Averaging ◽  
Benthic Species ◽  
Inference Models ◽  
Number Of Species ◽  
Chemical Conditions

Despite the overwhelming tendency in paleolimnology to use both planktonic and benthic diatoms when inferring open-water chemical conditions, it remains questionable whether all taxa are appropriate and necessary to construct useful inference models. We examined this question using a 75-lake training set from Quebec (Canada) to assess whether model performance is affected by the deletion of benthic species. Because benthic species are known to experience very different chemical conditions than their planktonic counterparts, we hypothesized that they would introduce undesirable noise in the calibration. Surprisingly, such important variables as pH, total phosphorus (TP), total nitrogen (TN), and dissolved organic carbon (DOC) were well predicted from weighted-averaging partial least square (WA-PLS) models based solely on benthic species. Similar results were obtained regardless of the depth of the lakes. Although the effective number of occurrence (N2) and the tolerance of species influenced the stability of the model residual error (jackknife), the number of species was the major factor responsible for the weaker inference models when based on planktonic diatoms alone. Indeed, when controlled for the number of species in WA-PLS models, individual planktonic diatom species showed superior predictive power over individual benthic species in inferring open-water chemical conditions.

scholarly journals Pengembangan Sensor FT-NIR Melalui Transformasi Wavelet Untuk Evaluasi Kadar Gula Mangga Gadung (Mangifera Indica)

2017 ◽  
Vol 2 (3) ◽  
pp. 276-293
Author(s):  
Muhammad Ikhram ◽  
Zulfahrizal Zulfahrizal ◽  
Agus Arip Munawar
Keyword(s):  
Prediction Model ◽  
Wavelet Transformation ◽  
Mangifera Indica ◽  
Sugar Content ◽  
Model Performance ◽  
Partial Least Square ◽  
Least Square ◽  
Sensor Technology ◽  
Baseline Correction ◽  
Mango Fruit

Abstrak: Sebelum adanya uji non-destruktif pada mangga untuk mengetahui kandungan kadar gula pada buah mangga selalu dilakukan metode destruktif yaitu dengan cara mangga diperas sari buahnya dan dilihat oBrix dengan menggunakan alat refraktometer. Penelitian ini bertujuan untuk mengembangkan instrument berbasis teknologi sensor FT-NIR melalui transformasi wavelet (wavelet segmentation) sehingga diharapkan dapat membantu mendeteksi cepat kualitas buah mangga. Penelitian ini menggunakan alat FT-NIR dengan sensor photodiode. Penelitian ini menggunakan model prediksi yang dibangun dengan menggunakan metode Partial least square dengan metode koreksi baseline correction. Setelah itu untuk mendeteksi data pencilan menggunakan metode analisa PCA dan hotelling T2 ellips sehingga data prediksi tidak ada noise (gangguan). Kemudian dilanjutkan dengan analisis laboratorium untuk mendapatkan nilai acuan dalam membangun model prediksi. Dalam membangun model prediksi Parameter statistika yang biasa digunakan untuk mengevaluasi model yang dihasilkan adalah Nilai Error (RMSEC), Nilai Koefisien Korelasi (r), Nilai Koefisien Determinasi (R2), dan RPD. Hasil penelitian menunjukan bahwa self developed FT-NIR mampu mendeteksi zat organik kadar gula dengan kisaran gelombang 2137 nm – 2333 nm, spektrum yang telah dikoreksi menggunakan baseline correction diperoleh nilai parameter statistiknya adalah R2 = 0,881, nilai r = 0,939, nilai RPD = 2,149, nilai error (RMSEC) sebesar 0,782. model yang dihasilkan adalah model prediksi yang bagus (good model performance) karena nilai RPD berada pada kisaran 2 - 2,5.Development of Fourier Transform Near InfraRed Spectroscopy (FT-NIR) Through Wavelet Transformation For Sugar Content Evaluation Mango Gadung (Mangifera Indica)Abstrack : Before the existence of non-destructive test on mango determine of sugar content in mango fruit always destructively by way of mango squeezed juice and seen Brix by using tool of refractometer. This research aims to develop intrument based on FT-NIR sensor technology through wavelet transformation (wavelet segmentation) so it is expected to help detect the quality of mango fruit fast. This research uses FT-NIR tool with photodiode sensor. This research uses prediction model which established by using partial least square method with correction method of baseline correction. Then proceed with laboratory analysis to obtain the reference value in building predictive model. In constructing the prediction model the usual statistical parameters used to evaluate the resulting model are error value (RMSEC), correlation coefficients (r), coefficient of determination (R2), and RPD. The results showed that self developed FT-NIR was able to detect organic subtance of sugar content with wave range 2137 nm - 2333 nm, the corrected spectra using baseline correction obtained statistic parameter value is R2 = 0,881, r = 0,939, value RPD = 2,149, error value (RMSEC) to 0,782. The model produced is a good model of performance (good model performance) because the value of RPD is in the range between 2 and 2,5.

scholarly journals Advancement of Individualized Head-Related Transfer Functions (HRTFs) in Perceiving the Spatialization Cues: Case Study for an Integrated HRTF Individualization Method

2019 ◽  
Vol 9 (9) ◽  
pp. 1867
Author(s):  
Lei Wang ◽  
Xiangyang Zeng ◽  
Xiyue Ma
Keyword(s):  
Transfer Functions ◽  
Objective Evaluation ◽  
Differential Method ◽  
Partial Least Square ◽  
Least Square ◽  
Experimental Results ◽  
Partial Least Square Regression ◽  
Sound Sources ◽  
Integrated Method

Head-related transfer function (HRTF), which varies across individuals at the same direction, has grabbed widespread attention in the field of acoustics and been used in many scenarios. In order to in-depth investigate the performance of individualized HRTFs on perceiving the spatialization cues, this study presents an integrated algorithm to obtain individualized HRTFs, and explores the advancement of such individualized HRTFs in perceiving the spatialization cues through two different binaural experiments. An integrated method for HRTF individualization on the use of Principle Component Analysis (PCA), Multiple Linear Regression (MLR) and Partial Least Square Regression (PLSR) was presented first. The objective evaluation was then made to verify the algorithmic effectiveness of that method. Next, two subjective experiments were conducted to explore the advancement of individualized HRTFs in perceiving the spatialization cues. One was auditory directional discrimination degree based on semantic differential method, in which the azimuth information of sound sources was told to the listeners before listening. The other was auditory localization, in which the azimuth information was not told to the listeners before listening. The corresponding statistical analyses for the subjective experimental results were made. All the experimental results support that individualized HRTFs obtained from the presented method achieve a preferable performance in perceiving the spatialization cues.

scholarly journals Error propagation in spectrometric functions of soil organic carbon

SOIL ◽  
2019 ◽  
Vol 5 (2) ◽  
pp. 275-288
Author(s):  
Monja Ellinger ◽  
Ines Merbach ◽  
Ulrike Werban ◽  
Mareike Ließ
Keyword(s):  
Organic Carbon ◽  
Soil Organic Carbon ◽  
Error Propagation ◽  
Field Experiments ◽  
Model Building ◽  
Model Performance ◽  
Partial Least Square ◽  
Least Square ◽  
Partial Least Square Regression ◽  
Measurement Protocol

Abstract. Soil organic carbon (SOC) plays a major role concerning chemical, physical, and biological soil properties and functions. To get a better understanding of how soil management affects the SOC content, the precise monitoring of SOC on long-term field experiments (LTFEs) is needed. Visible and near-infrared (Vis–NIR) reflectance spectrometry provides an inexpensive and fast opportunity to complement conventional SOC analysis and has often been used to predict SOC. For this study, 100 soil samples were collected at an LTFE in central Germany by two different sampling designs. SOC values ranged between 1.5 % and 2.9 %. Regression models were built using partial least square regression (PLSR). In order to build robust models, a nested repeated 5-fold group cross-validation (CV) approach was used, which comprised model tuning and evaluation. Various aspects that influence the obtained error measure were analysed and discussed. Four pre-processing methods were compared in order to extract information regarding SOC from the spectra. Finally, the best model performance which did not consider error propagation corresponded to a mean RMSEMV of 0.12 % SOC (R2=0.86). This model performance was impaired by ΔRMSEMV=0.04 % SOC while considering input data uncertainties (ΔR2=0.09), and by ΔRMSEMV=0.12 % SOC (ΔR2=0.17) considering an inappropriate pre-processing. The effect of the sampling design amounted to a ΔRMSEMV of 0.02 % SOC (ΔR2=0.05). Overall, we emphasize the necessity of transparent and precise documentation of the measurement protocol, the model building, and validation procedure in order to assess model performance in a comprehensive way and allow for a comparison between publications. The consideration of uncertainty propagation is essential when applying Vis–NIR spectrometry for soil monitoring.

scholarly journals A non-destructive method to quantify leaf starch content in red clover

2020 ◽  
Author(s):  
Lea Antonia Frey ◽  
Philipp Baumann ◽  
Helge Aasen ◽  
Bruno Studer ◽  
Roland Kölliker
Keyword(s):  
Energy Content ◽  
Starch Content ◽  
Model Performance ◽  
Red Clover ◽  
Partial Least Square ◽  
Least Square ◽  
Partial Least Square Regression ◽  
Test Set ◽  
Independent Test ◽  
Non Destructive

Abstract Background Grassland-based ruminant livestock production provides a sustainable alternative to intensive production systems relying on concentrated feeds. However, grassland-based roughage often lacks the energy content required to meet the productivity potential of modern livestock breeds. Forage legumes, such as red clover, with increased starch content could partly replace maize and cereal supplements. However, breeding for increased starch content requires efficient phenotyping methods. This study is unique in evaluating a non-destructive hyperspectral imaging approach to estimate leaf starch content in red clover for enabling efficient development of high starch red clover genotypes.Results We assessed prediction performance of partial least square regression models (PLSR) and validated model performance with an independent test set. Starch content of the training set ranged from 0.1 to 120.3 mg g -1 DW. The best cross-validated PLSR model explained 56% of the measured variation and yielded a root mean square error (RMSE) of 17 mg g -1 DW. Model performance decreased when applied to the independent test set (RMSE = 29 mg g -1 DW, R 2 = 0.36). Different filtering methods did not increase model performance.Conclusion The non-destructive spectral method presented here, provides a tool to detect large differences in leaf starch content of red clover. Breeding material can be sampled and selected according to their starch content without destroying the plant.

A high mountain pollen-altitude calibration set for palaeoclimatic use in the tropical Andes

The Holocene ◽  
2006 ◽  
Vol 16 (1) ◽  
pp. 105-117 ◽  
Author(s):  
Valenti Rull
Keyword(s):  
Altitudinal Gradient ◽  
Transfer Functions ◽  
Weighted Averaging ◽  
High Mountain ◽  
Mean Square ◽  
Tropical Andes ◽  
Training Set ◽  
Inference Models ◽  
Mountain Environments ◽  
M 12

The numerical relationship between modem pollen assemblages and altitude in high mountain environments from the northern Andes is analysed, in order to found inference models that allow estimating palaeoaltitudes and palaeotemperatures from past pollen records. The calibration set (DM) consists of a 50-sample altitudinal transect between-2300 and-4600 m altitude. The overall and individual pollen responses to altitude were tested by correspondence analysis (CA), generalized linear regression (HOF) and weighted averaging (WA). Transfer functions were derived by weighted averaging partial least squares (WA-PLS) regression. Overall, altitude is the main controlling factor for the composition of pollen assemblages, as shown by the high correlation between altitude and the first CA component (r =-0.88). Individually, around 35% of the 82 pollen taxa show a significant response to altitude through monotonic or unimodal functions. The best transfer function obtained has a good statistical performance, as shown by the determination coefficient (r2tck =0.78). The prediction power, as measured by the root mean square error of prediction (RMSEP), is of 256 m (12% of the total altitudinal gradient), which is equivalent to-1.5C. These parameters fall within the performance range of the inference models developed elsewhere using pollen and other biological proxies. It is concluded that the DM training set is useful to reconstruct Pleistocene and major Holocene palaeoclimatic trends. This study demonstrates the suitability of establishing reliable transfer functions for palaeoclimatic estimation in the highest altitudes of the tropical Andes, and encourages their continued improvement.

Tolerance of near Infrared Calibrations to Temperature Variations; A Practical Evaluation

2000 ◽  
Vol 8 (2) ◽  
pp. 125-132 ◽  
Author(s):  
W.G. Hansen ◽  
S.C.C. Wiedemann ◽  
M. Snieder ◽  
V.A.L. Wortel
Keyword(s):  
Near Infrared ◽  
Model Performance ◽  
Partial Least Square ◽  
Least Square ◽  
Temperature Variations ◽  
Precise Control ◽  
Practical Evaluation ◽  
Temperature Variable ◽  
Value Determination ◽  
Model Robustness

Process samples of esters have been analysed by transmittance near infrared (NIR) at temperatures between 60 and 70°C (±0.2). Apart from density changes, these small temperature variations affect molecular associations by H-bonding. Partial Least Square (PLS) models based on the first OH overtone (1350–1500nm) have been made for hydroxyl value determination, including implicitly the temperature variable. The sensitivity of these NIR calibrations to temperature has been evaluated by an analysis of variance and the “Taguchi principle”, using both average model performance and model variance. An accurate and precise control of the sample temperature prior to scanning leads to the lowest prediction error. When temperature fluctuations can not be avoided, introduction of temperature variance in the calibration set can improve the model robustness; this strategy is only beneficial if temperature range, temperature distribution and number of PLS factors have been carefully optimised.

scholarly journals A non-destructive method to quantify leaf starch content in red clover

2020 ◽  
Author(s):  
Lea Antonia Frey ◽  
Philipp Baumann ◽  
Helge Aasen ◽  
Bruno Studer ◽  
Roland Kölliker
Keyword(s):  
Energy Content ◽  
Starch Content ◽  
Model Performance ◽  
Red Clover ◽  
Partial Least Square ◽  
Least Square ◽  
Partial Least Square Regression ◽  
Test Set ◽  
Independent Test ◽  
Non Destructive

Abstract Background: Grassland-based ruminant livestock production provides a sustainable alternative to intensive production systems relying on concentrated feeds. However, grassland-based roughage often lacks the energy content required to meet the productivity potential of modern livestock breeds. Forage legumes, such as red clover, with increased starch content could partly replace maize and cereal supplements. However, breeding for increased starch content requires efficient phenotyping methods. This study is unique in evaluating a non-destructive hyperspectral imaging approach to estimate leaf starch content in red clover for enabling efficient development of high starch red clover genotypes. Results: We assessed prediction performance of partial least square regression models (PLSR) using cross-validation, and validated model performance with an independent test set. Starch content of the training set ranged from 0.1 to 120.3 mg g-1 DW. The best cross-validated PLSR model explained 56% of the measured variation and yielded a root mean square error (RMSE) of 17 mg g-1 DW. Model performance decreased when applying the trained model on the independent test set (RMSE = 29 mg g-1 DW, R2 = 0.36). Different variable selection methods did not increase model performance. Conclusion: The non-destructive spectral method presented here, provides a tool to detect large differences in leaf starch content of red clover. The major benefit of the method is that it can be repeatedly applied to the same plants, thus providing a means to follow starch concentrations over time and over a broad range of environments.

scholarly journals Comprehensive model optimization in pulp quality prediction: a machine learning approach

2017 ◽  
Author(s):  
Chin-Wei Huang ◽  
Luc Baron ◽  
Marek Balazinski ◽  
Sofiane Achiche
Keyword(s):  
Machine Learning ◽  
Genetic Algorithm ◽  
Fuzzy Inference ◽  
Model Performance ◽  
Partial Least Square ◽  
Least Square ◽  
Partial Least Square Regression ◽  
Special Importance ◽  
Quality Prediction ◽  
Inference System

Feature selection in machine learning is of great interest since it is reckoned as creating more efficient predictive models in several engineering domains. It is even of special importance in the pulp and paper transformation industry as the knowledge of this particular process is generally very limited. In this paper, we first compared the performance of rule-based genetic algorithm and that of adaptive neuro-fuzzy inference system; the latter is found to be more precise in predicting the pulp quality. We then combined several data mining algorithms such as genetic algorithm-partial least square regression, along with other statistical methods, to explore the relevance of all the potential variables that could be used to predict the pulp ISO brightness, an important property that is usually linked to model performance and hence pulp quality prediction. A few highly relevant variables are thereby determined, and the full set of 79 variables obtained from a Chip Management System was trimmed down to an optimized combination of 3 inputs depending on their relevancy. Peroxide charge (P), average luminance (L) and hue (H) were chosen as the optimal subset to describe the ISO brightness of the pulp and the model was simplified without losing much of its accuracy. Finally, we derived the numbers of membership functions for each variable to further refine the fuzzy logic-based prediction model. The error then reached 2.18%. The loss on accuracy was compensated by adjusting to the fittest membership function numbers

Quantitative reconstruction of past salinity variations in African lakes: assessment of chironomid-based inference models (Insecta: Diptera) in space and time

2004 ◽  
Vol 61 (6) ◽  
pp. 986-998 ◽  
Author(s):  
Dirk Verschuren ◽  
Brian F Cumming ◽  
Kathleen R Laird
Keyword(s):  
Surface Waters ◽  
Model Performance ◽  
Weighted Averaging ◽  
Quantitative Reconstruction ◽  
Inference Models ◽  
African Lakes ◽  
Mean Squared Errors ◽  
Calibration Models ◽  
Fossil Data ◽  
Salinity Variations

Faunal records of 20 common midge species (Diptera: Chironomidae) in 32 African surface waters with salinities ranging from 20 to 41 000 µS·cm–1 were used to develop inference models for quantitative reconstruction of past salinity variations from larval chironomid fossils preserved in lake sediments. Weighted-averaging regression and calibration models using presence–absence data (P/A) and presence–absence data with tolerance down-weighting (P/Atol) produced bootstrapped coefficients of determination (r2) of 0.78 and 0.81, respectively, and root mean squared errors (RMSE) of prediction of 0.42 and 0.39 log conductivity units. Historical conductivity data from African lakes are scarce. Therefore, model performance was tested in time by comparing chironomid-inferred conductivity estimates with the corresponding diatom-inferred estimates in sediment records of two fluctuating lakes in the Rift Valley of Kenya. A hybrid procedure in which presence–absence calibration models were applied to abundance-weighted fossil data yielded significantly higher correlation between chironomid- and diatom-inferred time series (Lake Oloidien AD 1880–1991, r2 = 0.76–0.78; Crescent Island Crater AD 900–1993, r2 = 0.56–0.61) than by applying the same models to presence–absence fossil data (r2 = 0.47–0.56 and 0.26–0.42, respectively). Overall, model performance confirms that Chironomidae are valuable bioindicators for natural and man-made changes in the water balance of African lakes.

scholarly journals Comprehensive model optimization in pulp quality prediction: a machine learning approach

2017 ◽  
Author(s):  
Chin-Wei Huang ◽  
Luc Baron ◽  
Marek Balazinski ◽  
Sofiane Achiche
Keyword(s):  
Machine Learning ◽  
Genetic Algorithm ◽  
Fuzzy Inference ◽  
Model Performance ◽  
Partial Least Square ◽  
Least Square ◽  
Partial Least Square Regression ◽  
Special Importance ◽  
Quality Prediction ◽  
Inference System

Feature selection in machine learning is of great interest since it is reckoned as creating more efficient predictive models in several engineering domains. It is even of special importance in the pulp and paper transformation industry as the knowledge of this particular process is generally very limited. In this paper, we first compared the performance of rule-based genetic algorithm and that of adaptive neuro-fuzzy inference system; the latter is found to be more precise in predicting the pulp quality. We then combined several data mining algorithms such as genetic algorithm-partial least square regression, along with other statistical methods, to explore the relevance of all the potential variables that could be used to predict the pulp ISO brightness, an important property that is usually linked to model performance and hence pulp quality prediction. A few highly relevant variables are thereby determined, and the full set of 79 variables obtained from a Chip Management System was trimmed down to an optimized combination of 3 inputs depending on their relevancy. Peroxide charge (P), average luminance (L) and hue (H) were chosen as the optimal subset to describe the ISO brightness of the pulp and the model was simplified without losing much of its accuracy. Finally, we derived the numbers of membership functions for each variable to further refine the fuzzy logic-based prediction model. The error then reached 2.18%. The loss on accuracy was compensated by adjusting to the fittest membership function numbers

Sign in / Sign up

Export Citation Format

Share Document