Ridge and PLS based rational function regression

Taavitsainen, Veli‐Matti

doi:10.1002/cem.1328

Cited by 13 publications

(6 citation statements)

References 16 publications

(19 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to studying the usability of handheld instruments in analyzing woodchip samples, our aim was to investigate how much the prediction performance could be improved by using a nonlinear method (RRR technique) rather than a linear one (PLS technique). RRR technique was chosen among the different nonlinear regression techniques mainly for two reasons: It is relatively new, and being based on linearization, it is fast compared to many other nonlinear methods 18 . By the results, it is clear that in average RRR performs (statistically) significantly better than PLS, although with raw data, in spite of the statistical significance, the practical significance of the difference is questionable.…”

Section: Discussionmentioning

confidence: 99%

“…Thanks to chemometric techniques, the amount of light absorbed can give quantitative information related to the moisture content of the material. In detail, two regression methods have been used in order to predict the moisture content of woodchip samples, that is, Partial Least Squares (PLS) 17 and Rational function Ridge Regression (RRR) 18,19 . PLS is perhaps the most used approach for prediction using spectral data, and it is based on the relationship between Y variable, that is, the parameter to be predicted, and the X variable, that is, the spectral values.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Comparative study between Partial Least Squares and Rational function Ridge Regression models for the prediction of moisture content of woodchip samples using a handheld spectrophotometer

Mancini

Taavitsainen

Toscano

2021

Journal of Chemometrics

View full text Add to dashboard Cite

The use of woodchip for energy use is expected to increase in the next years because of the European targets for mitigating climate change and reducing greenhouse gas emissions. The technical standard EN ISO 17225‐4 determines the woodchip quality classes based on different chemical–physical parameters. Among them, moisture content is one of the most important, and its real‐time monitoring could improve the product quality, increase combustion efficiency, and consequently provide a potential decrease of pollutant emissions. Although the lab procedure to determine moisture content is quite simple, it is too long compared to the real needs of power plant operators. A rapid and simple solution may be represented by near‐infrared spectroscopy coupled with chemometric techniques. In detail, two regression methods, that is, Partial Least Squares (PLS) and Rational function Ridge Regression (RRR), have been compared in order to develop a good model for the prediction of moisture content of woodchip samples as arrived at the lab or directly in the power plant. In addition, the prediction performance has been studied as a function of the number of the spectral measurements of the sample. The results showed that handheld instruments could give reliable results by taking enough replicated spectra. According to the estimated measurement uncertainty of moisture measurements, both methods suffer from lack of fit, and the performance could be better if all unknown sources of errors could be eliminated. In general, RRR performs significantly better than PLS, although, as with many nonlinear methods, the risk for outliers in predictions with RRR is higher than with PLS. This could be overpassed performing predictions with both methods, and in cases of big differences, PLS prediction could be chosen.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Comparative study between Partial Least Squares and Rational function Ridge Regression models for the prediction of moisture content of woodchip samples using a handheld spectrophotometer

Mancini

Taavitsainen

Toscano

2021

Journal of Chemometrics

View full text Add to dashboard Cite

show abstract

“…To alleviate this issue, in [13] the authors proposed an extension to this representation called Transformation-Interaction-Rational representation (TIR). In short, it simply combines two IT expressions as a rational polynomial regression model [23][24][25] and apply an invertible function to the resulting value:…”

Section: Transformation-interaction-rationalmentioning

confidence: 99%

Multi-Objective Optimization with Transformation-Interaction-Rational Representation for Symbolic Regression

França

2022

Preprint

View full text Add to dashboard Cite

The Transformation-Interaction-Rational is a representation for symbolic regression that limits the search space of function forms to well structured functions enabling the use of efficient numerical coefficient determination. The performance of using Genetic Programming with this representation was substantially better than with its predecessor (Interaction-Transformation) while ranking close to the state-of-the-art on a contemporary Symbolic Regression benchmark. In this same benchmark, we observed that the performance could be further improved with an additional selective pressure on smaller datasets for the expression size, thus creating smaller expressions and generalizing better. This was verified with a penalization term applied to the fitness measure together with heuristic rules to determine what is a small dataset. Possibly a better strategy is to adopt a multi-objective search to find a set of non-dominated solutions taking accuracy and model size in consideration. This will automatically stimulate the preference for simpler models whenever two equally accurate model exists, possibly the need for heuristic rules and penalty functions. In this paper, we extend the evolution library used by TIR to support multi-objective optimization, specifically the NSGA-II algorithm, and apply that to the same benchmark. The results show that the use of multi-objective optimization improves the results on a selection of the benchmarks increasing the overall rank of the algorithm to a second place in the benchmark. However, this strategy is still sensitive to overfitting on small data sets.

show abstract

“…The following is a program developed in R statistical software, which can be used for processing the spectral data using various calibration models. The code will be modified to implement several variations of the principal components regression (PCR) and the partial least squares (PLS) methods proposed in the literature [5][6][7][8][9][10][11][12]. The results of all these methods will be compared against conventional PCR/PLS and GC measurements to select the best method for our purpose.…”

Section: Preliminary Experiments With Producer Gasmentioning

confidence: 99%

Integrated Advanced Reciprocating Internal Combustion Engine System for Increased Utilization of Gaseous Opportunity Fuels

Pratapas

Zelepouga

Gnatenko

et al. 2013

View full text Add to dashboard Cite

Ridge and PLS based rational function regression

Cited by 13 publications

References 16 publications

Comparative study between Partial Least Squares and Rational function Ridge Regression models for the prediction of moisture content of woodchip samples using a handheld spectrophotometer

Comparative study between Partial Least Squares and Rational function Ridge Regression models for the prediction of moisture content of woodchip samples using a handheld spectrophotometer

Multi-Objective Optimization with Transformation-Interaction-Rational Representation for Symbolic Regression

Integrated Advanced Reciprocating Internal Combustion Engine System for Increased Utilization of Gaseous Opportunity Fuels

Contact Info

Product

Resources

About