A New Variable Selection Method Based on Mutual Information Maximization by Replacing Collinear Variables for Nonlinear Quantitative Structure-Property Relationship Models

Ghasemi, Jahan B.; Zolfonoun, Ehsan

doi:10.5012/bkcs.2012.33.5.1527

Cited by 4 publications

(2 citation statements)

References 42 publications

(37 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The experimental values of DT50 can be compared to the predicted values obtained with the quantitative structure property relationship models, which were developed based on theoretically derived molecular descriptors. Ghasemi and Zolfonoun (2012) estimated a DT50 value for glyphosate of 18.1 days with the support vector machines (SVM) approach and D r a f t of 12.0 days with the radial basis function neural network approach. A DT50 of 9.5 days was predicted by Samghani and Fatemi (2016) via the SVM approach.…”

Section: R a F Tmentioning

confidence: 99%

Review of studies analysing glyphosate and aminomethylphosphonic acid (AMPA) occurrence in groundwater

2022

View full text Add to dashboard Cite

The herbicide glyphosate (N-phosphonomethylglycine) has been the most commonly applied herbicide worldwide for the past 40 years. This review aims to present and discuss the state of knowledge concerning groundwater contamination by glyphosate and its metabolite, aminomethylphosphonic acid (AMPA). A dataset of 48 reports and articles reporting glyphosate and AMPA detection in groundwater worldwide was constructed from published literature. Specific attention was given to the analysis of available studies on glyphosate and AMPA transport in groundwater to investigate groundwater contamination factors. The information presented in this review highlights detectable groundwater contamination in several countries globally, with many cases exceeding European groundwater quality standards. Both agricultural and non-agricultural applications of glyphosate can be significant sources of pollution, and proximity to agricultural fields is a significant risk factor. AMPA has two primary sources, glyphosate and amino polyphosphates, but very little information is available about the origin of AMPA detected in groundwater. Glyphosate and AMPA transport to groundwater can be significant in well-structured soils rich in macropores, where it generally occurs via preferential flow. An inverse relationship was highlighted between the groundwater depth and glyphosate and AMPA occurrence and concentration in groundwater; however, some discrepancies among the studies were identified. Heavy rainfall shortly after glyphosate application poses a high risk of its transport to groundwater. Seasonal trends in glyphosate and AMPA in groundwater, related to application timing, agronomic practices, and weather conditions, were also observed. Finally, risk-management measures are proposed.

show abstract

Section: R a F Tmentioning

confidence: 99%

Review of studies analysing glyphosate and aminomethylphosphonic acid (AMPA) occurrence in groundwater

2022

View full text Add to dashboard Cite

show abstract

“…There are a number of works related to variable selection based on mutual informa tion, e.g. [7]- [9]. Mutual information is based on the concept of entropy of a random variable.…”

Section: A Pcamentioning

confidence: 99%

Information-theoretic approach to variable selection in predictive models applied to paper machine data

Ivannikova

Hämäläinen

Luostarinen

2013

2013 IEEE Symposium on Computers and Communications (ISCC)

View full text Add to dashboard Cite

This paper presents an information-theoretic ap proach to variable selection for prediction of laboratory mea surements of paper quality. Along with a well-known Principal Component Analysis we considered techniques for variable selec tion based on the classical Shannon Mutual Information and a novel Maximal Information Coefficient. A multilayer perceptron neural model was used to predict quality measurements and compare feature selection techniques. The suggested approach was tested on real industrial data obtained form a pilot paper machine. The presented results show that information-theoretic techniques perform better compared to Principal Component Analysis, providing higher accuracy results.

show abstract

Determination of boiling points of azeotropic mixtures using quantitative structure–property relationship (QSPR) strategy

Zare-Shahabadi

Lotfi

Gandomani

et al. 2013

Journal of Molecular Liquids

View full text Add to dashboard Cite

A New Variable Selection Method Based on Mutual Information Maximization by Replacing Collinear Variables for Nonlinear Quantitative Structure-Property Relationship Models

Cited by 4 publications

References 42 publications

Review of studies analysing glyphosate and aminomethylphosphonic acid (AMPA) occurrence in groundwater

Review of studies analysing glyphosate and aminomethylphosphonic acid (AMPA) occurrence in groundwater

Information-theoretic approach to variable selection in predictive models applied to paper machine data

Determination of boiling points of azeotropic mixtures using quantitative structure–property relationship (QSPR) strategy

Contact Info

Product

Resources

About