Comparison of multivariate models and variable selection algorithms for rapid analysis of the chemical composition of field crops
This study evaluates the use of visible and near-infrared spectroscopy for rapid prediction of total carbon, total nitrogen, and total phosphorus concentrations in field crop samples. Two multivariate models (partial least squares regression and support vector machine regression) were compared. In a...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2019-07-01
|
| Series: | Acta Agriculturae Scandinavica. Section B, Soil and Plant Science |
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1080/09064710.2019.1600012 |
| _version_ | 1797684407340892160 |
|---|---|
| author | Shengxiang Xu Meiyan Wang Xuezheng Shi |
| author_facet | Shengxiang Xu Meiyan Wang Xuezheng Shi |
| author_sort | Shengxiang Xu |
| collection | DOAJ |
| description | This study evaluates the use of visible and near-infrared spectroscopy for rapid prediction of total carbon, total nitrogen, and total phosphorus concentrations in field crop samples. Two multivariate models (partial least squares regression and support vector machine regression) were compared. In addition, four spectral variable selection algorithms (competitive adaptive reweighted sampling, genetic algorithm, uninformative variable elimination, and variable importance for projection) were applied with support vector machine regression to determine the most accurate predictions. The results showed that support vector machine regression performed better than partial least squares regression for predicting the three chemical compositions. The combination of competitive adaptive reweighted sampling and support vector machine regression outperformed the other models for the predictions of total carbon and total nitrogen with high coefficients of determination of 0.91 and 0.90, respectively. For the determination of total phosphorus, the prediction accuracy of competitive adaptive reweighted sampling was comparable with the best result obtained from genetic algorithm with the coefficients of determination of 0.73 and 0.77, respectively. In conclusion, the support vector machine regression combined with competitive adaptive reweighted sampling has great potential to accurately determine the chemical composition of field crops using the visible and near-infrared spectroscopy. |
| first_indexed | 2024-03-12T00:29:12Z |
| format | Article |
| id | doaj.art-6f59cc096dad4376967ced5a1b1669f4 |
| institution | Directory Open Access Journal |
| issn | 0906-4710 1651-1913 |
| language | English |
| last_indexed | 2024-03-12T00:29:12Z |
| publishDate | 2019-07-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Acta Agriculturae Scandinavica. Section B, Soil and Plant Science |
| spelling | doaj.art-6f59cc096dad4376967ced5a1b1669f42023-09-15T10:26:24ZengTaylor & Francis GroupActa Agriculturae Scandinavica. Section B, Soil and Plant Science0906-47101651-19132019-07-0169545246410.1080/09064710.2019.16000121600012Comparison of multivariate models and variable selection algorithms for rapid analysis of the chemical composition of field cropsShengxiang Xu0Meiyan Wang1Xuezheng Shi2State Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of SciencesState Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of SciencesState Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of SciencesThis study evaluates the use of visible and near-infrared spectroscopy for rapid prediction of total carbon, total nitrogen, and total phosphorus concentrations in field crop samples. Two multivariate models (partial least squares regression and support vector machine regression) were compared. In addition, four spectral variable selection algorithms (competitive adaptive reweighted sampling, genetic algorithm, uninformative variable elimination, and variable importance for projection) were applied with support vector machine regression to determine the most accurate predictions. The results showed that support vector machine regression performed better than partial least squares regression for predicting the three chemical compositions. The combination of competitive adaptive reweighted sampling and support vector machine regression outperformed the other models for the predictions of total carbon and total nitrogen with high coefficients of determination of 0.91 and 0.90, respectively. For the determination of total phosphorus, the prediction accuracy of competitive adaptive reweighted sampling was comparable with the best result obtained from genetic algorithm with the coefficients of determination of 0.73 and 0.77, respectively. In conclusion, the support vector machine regression combined with competitive adaptive reweighted sampling has great potential to accurately determine the chemical composition of field crops using the visible and near-infrared spectroscopy.http://dx.doi.org/10.1080/09064710.2019.1600012model calibrationsupport vector machine regressiontotal carbontotal nitrogentotal phosphorusvisible and near-infrared spectroscopy |
| spellingShingle | Shengxiang Xu Meiyan Wang Xuezheng Shi Comparison of multivariate models and variable selection algorithms for rapid analysis of the chemical composition of field crops Acta Agriculturae Scandinavica. Section B, Soil and Plant Science model calibration support vector machine regression total carbon total nitrogen total phosphorus visible and near-infrared spectroscopy |
| title | Comparison of multivariate models and variable selection algorithms for rapid analysis of the chemical composition of field crops |
| title_full | Comparison of multivariate models and variable selection algorithms for rapid analysis of the chemical composition of field crops |
| title_fullStr | Comparison of multivariate models and variable selection algorithms for rapid analysis of the chemical composition of field crops |
| title_full_unstemmed | Comparison of multivariate models and variable selection algorithms for rapid analysis of the chemical composition of field crops |
| title_short | Comparison of multivariate models and variable selection algorithms for rapid analysis of the chemical composition of field crops |
| title_sort | comparison of multivariate models and variable selection algorithms for rapid analysis of the chemical composition of field crops |
| topic | model calibration support vector machine regression total carbon total nitrogen total phosphorus visible and near-infrared spectroscopy |
| url | http://dx.doi.org/10.1080/09064710.2019.1600012 |
| work_keys_str_mv | AT shengxiangxu comparisonofmultivariatemodelsandvariableselectionalgorithmsforrapidanalysisofthechemicalcompositionoffieldcrops AT meiyanwang comparisonofmultivariatemodelsandvariableselectionalgorithmsforrapidanalysisofthechemicalcompositionoffieldcrops AT xuezhengshi comparisonofmultivariatemodelsandvariableselectionalgorithmsforrapidanalysisofthechemicalcompositionoffieldcrops |