zhpsЖурнал прикладной спектроскопииZhurnal Prikladnoii Spektroskopii0514-7506B. I. Stepanov Institute of Physics of the National Academy of Scienceszhps-1194Research ArticleАННОТАЦИИ АНГЛОЯЗЫЧНЫХ СТАТЕЙABSTRACTS ENGLISH-LANGUAGE ARTICLESАнсамблевый метод обучения на основе суммирования для прогнозирования доступного азота в почве с помощью портативного микроспектрометра ближнего инфракрасного диапазонаA Stacking-Based Ensemble Learning Method for Available Nitrogen Soil Prediction with a Handheld Micronear-Infrared SpectrometerWanM.WanM.

Хэфэй

Hefei

JinX.JinX.

Хэфэй

Hefei

jinxiu123@ahau.edu.cnHanY.HanY.

Хэфэй

Hefei

WangL.WangL.

Хэфэй

Hefei

LiS.LiS.

Хэфэй

Hefei

RaoY.RaoY.

Хэфэй

Hefei

ZhangX.ZhangX.

Хэфэй

Hefei

GaoQ.GaoQ.

Хэфэй

Hefei

Аньхойский сельскохозяйственный университет; Колледж информации и компьютерных наук Аньхойского сельскохозяйственного университетаAnhui Province Key Laboratory of Smart Agricultural Technology and Equipment, Anhui Agriculture University; College of Information and Computer Science at Anhui Agricultural University202202122022896907907Copyright © Wan M., Jin X., Han Y., Wang L., Li S., Rao Y., Zhang X., Gao Q., 20222022Wan M., Jin X., Han Y., Wang L., Li S., Rao Y., Zhang X., Gao Q.Wan M., Jin X., Han Y., Wang L., Li S., Rao Y., Zhang X., Gao Q.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://zhps.ejournal.by/jour/article/view/1194

Для разработки надежной и точной модели доступного в почве азота используются спектроскопия в ближнем ИК-диапазоне (БИК) почвы и данные об азоте. Спектральная отражательная способность образцов почвы изучена в диапазоне 900—1700 нм с помощью девяти методов предварительной обработки с использованием портативного микроБИК-спектрометра. Экстремальное повышение градиента (XGBoost), категориальное повышение (CatBoost), машина повышения градиента света (LightGBM) и случайный лес, которые представляют собой древовидные алгоритмы, сложены в виде базовых моделей. Линейная регрессия используется в метамодели для определения уникального шаблона обучения базовой модели. Результаты показывают, что диапазон и характеристики спектров можно использовать для соответствующих прогнозов, а спектры микро-БИК меняются при различных предварительных обработках. Многослойная модель обеспечивает наилучшую производительность среди всех протестированных моделей. Коэффициент детерминации R2 = 0.942, относительная процентная разница RPD = 4.192 с коррекцией Савицкого–Голея и мультипликативным рассеянием. 

Soil-available nitrogen is a vital index related to the growth and development of crops. The real-time and non-destructive detection of the soil-available nitrogen content based on near-infrared (NIR) spectroscopy could improve the accurate management of crop nutrients. In this manuscript, soil NIR spectroscopy and available nitrogen data are used in a stacked framework to develop a reliable and accurate soilavailable nitrogen model. The spectral reflectance of the soil samples was collected in the 900 to 1700 nm band with nine pre-processing methods using a handheld micronear-infrared spectrometer. The stacking framework of this manuscript has two layers. Extreme gradient boosting (XGBoost), categorical boosting (CatBoost), a light gradient boosting machine (LightGBM) and a random forest, which are tree-based algorithms, are stacked as base models in the first layer. In the second layer, linear regression is employed in a meta-model to identify the unique learning pattern of the base model. The results show that the range and characteristics of the spectra can be used to make relevant predictions, and the micro-NIR spectra are variable under different pre-treatments. In addition, the stacked model achieves the best performance of all the models tested. Notably, the coefficient of determination (R2) is 0.942, and the relative percent difference is 4.192 with Savitzky–Golay and multiplicative scatter correction. This manuscript presents an efficient method for predicting soil-available nitrogen levels with a handheld micronear-infrared spectrometer. 

ближний инфракрасный диапазондоступный азот в почвепортативный спектрометрпредварительная обработкаансамблевый метод обученияnear-infrared spectrumsoil-available nitrogenhandheld micronear-infrared spectrometerpretreatmentstacked ensemble learning methodWe are very grateful for the support of Natural Science Foundation of Anhui Province (No. 2008085MF203), Anhui Provincial Natural Science Project (No. KJ2019Z20), Anhui Provincial Natural Science Project (No. KJ2019A0212), Anhui Province Key Research and Development Program Project (No. 1804a07020108), Anhui Province Key Research and Development Program Project (No. 201904a06020056).ReferencesX. Mu, Y. Chen, Plant Physiol. Biochem., 158, 76–82 (2021).X. Mu, Y. Chen, Plant Physiol. Biochem., 158, 76–82 (2021).X. Wang, J. Fan, Y. Xing, G. Xu, H. Wang, J. Deng, Y. Wang, F. Zhang, P. Li, Z. Li, Adv. Agronomy, 153, 121–173 (2019).X. Wang, J. Fan, Y. Xing, G. Xu, H. Wang, J. Deng, Y. Wang, F. Zhang, P. Li, Z. Li, Adv. Agronomy, 153, 121–173 (2019).T. Terhoeven-Urselmans, H. Schmidt, R. G. Joergensen, B. Ludwig, Soil Biol. Biochem., 40, N 5, 1178–1188 (2008).T. Terhoeven-Urselmans, H. Schmidt, R. G. Joergensen, B. Ludwig, Soil Biol. Biochem., 40, N 5, 1178–1188 (2008).R. V. Rossel, R. Webster, Eur. J. Soil Sci., 63, N 6, 848–860 (2012).R. V. Rossel, R. Webster, Eur. J. Soil Sci., 63, N 6, 848–860 (2012).R. V. Rossel, S. R. Cattle, A. Ortega, Y. Fouad, Geoderma, 150, N 3-4, 253–266 (2009).R. V. Rossel, S. R. Cattle, A. Ortega, Y. Fouad, Geoderma, 150, N 3-4, 253–266 (2009).B. Stenberg, R. A. V. Rossel, A. M. Mouazen, J. Wetterlind, Adv. Agronomy, 107, 163–215 (2010).B. Stenberg, R. A. V. Rossel, A. M. Mouazen, J. Wetterlind, Adv. Agronomy, 107, 163–215 (2010).S. Nawar, A. M. Mouazen, Sensors, 17, N 10, 2428 (2017).S. Nawar, A. M. Mouazen, Sensors, 17, N 10, 2428 (2017).T. Leng, F. Li, Y. Chen, L. Tang, J. Xie, Q. Yu, Meat Sci., 180, 108559 (2021).T. Leng, F. Li, Y. Chen, L. Tang, J. Xie, Q. Yu, Meat Sci., 180, 108559 (2021).M. Knadel, L. W. de Jonge, M. Tuller, H. U. Rehman, P. W. Jensen, P. Moldrup, M. H. Greve, E. Arthur, Vadose Zone J., 19, N 1, e20007 (2020).M. Knadel, L. W. de Jonge, M. Tuller, H. U. Rehman, P. W. Jensen, P. Moldrup, M. H. Greve, E. Arthur, Vadose Zone J., 19, N 1, e20007 (2020).V. Ulissi, F. Antonucci, P. Benincasa, M. Farneselli, G. Tosti, M. Guiducci, F. Tei, C. Costa, F. Pallottino, L. Pari, Sensors, 11, N 6, 6411–6424 (2011).V. Ulissi, F. Antonucci, P. Benincasa, M. Farneselli, G. Tosti, M. Guiducci, F. Tei, C. Costa, F. Pallottino, L. Pari, Sensors, 11, N 6, 6411–6424 (2011).Y. Shao, Y. He, Soil Res., 49, N 2, 166–172 (2011).Y. Shao, Y. He, Soil Res., 49, N 2, 166–172 (2011).J. Tang, J. Liang, C. Han, Z. Li, H. Huang, Accident Analysis & Prevention, 122, 226–238 (2019).J. Tang, J. Liang, C. Han, Z. Li, H. Huang, Accident Analysis & Prevention, 122, 226–238 (2019).H.-C. Yi, Z.-H. You, M.-N. Wang, Z.-H. Guo, Y.-B. Wang, J.-R. Zhou, BMC Bioinformatics, 21, N 1, 1–10 (2020).H.-C. Yi, Z.-H. You, M.-N. Wang, Z.-H. Guo, Y.-B. Wang, J.-R. Zhou, BMC Bioinformatics, 21, N 1, 1–10 (2020).F. Liu, R. Zhao, L. Shi, arXiv preprint arXiv, 2103, 13124 (2021).F. Liu, R. Zhao, L. Shi, arXiv preprint arXiv, 2103, 13124 (2021).Å. Rinnan, F. V. D. Berg, S. B. Engelsen, TrAC Trends Anal. Chem., 28, N 10, 1201–1222 (2009).Å. Rinnan, F. V. D. Berg, S. B. Engelsen, TrAC Trends Anal. Chem., 28, N 10, 1201–1222 (2009).A. Savitzky, M. J. Golay, Anal. Chem., 36, N 8, 1627–1639 (1964).A. Savitzky, M. J. Golay, Anal. Chem., 36, N 8, 1627–1639 (1964).S. Nawar, H. Buddenbaum, J. Hill, J. Kozak, A. M. Mouazen, Soil and Tillage Res., 155, 510–522 (2016).S. Nawar, H. Buddenbaum, J. Hill, J. Kozak, A. M. Mouazen, Soil and Tillage Res., 155, 510–522 (2016).R. J. Barnes, M. S. Dhanoa, S. J. Lister, Appl. Spectr., 43, N 5, 772–777 (1989).R. J. Barnes, M. S. Dhanoa, S. J. Lister, Appl. Spectr., 43, N 5, 772–777 (1989).T. Isaksson, T. Næs, Appl. Spectr., 42, N 7, 1273–1284 (1988).T. Isaksson, T. Næs, Appl. Spectr., 42, N 7, 1273–1284 (1988).A. Peirs, A. Schenk, B. M. Nicolaı̈, Postharvest Biology and Technology, 35, N 1, 1–13 (2005).A. Peirs, A. Schenk, B. M. Nicolaı̈, Postharvest Biology and Technology, 35, N 1, 1–13 (2005).M. S. Askari, J. Cui, S. M. O’Rourke, N. M. Holden, Soil and Tillage Res., 146, 108–117 (2015).M. S. Askari, J. Cui, S. M. O’Rourke, N. M. Holden, Soil and Tillage Res., 146, 108–117 (2015).H. Liang, M. Zhang, C. Gao, Y. Zhao, Sensors, 18, N 6, 1963 (2018).H. Liang, M. Zhang, C. Gao, Y. Zhao, Sensors, 18, N 6, 1963 (2018).X. Jin, L. Wang, W. Zheng, X. Zhang, L. Liu, S. Li, Y. Rao, J. Xuan, Measurement, 110553 (2021).X. Jin, L. Wang, W. Zheng, X. Zhang, L. Liu, S. Li, Y. Rao, J. Xuan, Measurement, 110553 (2021).A. Gholizadeh, L. Borůvka, M. M. Saberioon, J. Kozak, R. Vašát, K. Němeček, Soil and Water Res., 10, N 4, 218–227 (2015).A. Gholizadeh, L. Borůvka, M. M. Saberioon, J. Kozak, R. Vašát, K. Němeček, Soil and Water Res., 10, N 4, 218–227 (2015).Y. Sun, M. Yuan, X. Liu, M. Su, L. Wang, Y. Zeng, H. Zang, L. Nie, Microchem. J., 159, 105492 (2020).Y. Sun, M. Yuan, X. Liu, M. Su, L. Wang, Y. Zeng, H. Zang, L. Nie, Microchem. J., 159, 105492 (2020).J. Duckworth, Near‐Infrared Spectrosc. Agric., 44, 113–132 (2004).J. Duckworth, Near‐Infrared Spectrosc. Agric., 44, 113–132 (2004).M. S. Dhanoa, S. J. Lister, R. Sanderson, R. J. Barnes, J. Near Infrared Spectrosc., 2, N 1, 43–47 (1994).M. S. Dhanoa, S. J. Lister, R. Sanderson, R. J. Barnes, J. Near Infrared Spectrosc., 2, N 1, 43–47 (1994).B. M. Nicolai, K. Beullens, E. Bobelyn, A. Peirs, W. Saeys, K. I. Theron, J. Lammertyn, Postharvest Biology and Technology, 46, N 2, 99–118 (2007).B. M. Nicolai, K. Beullens, E. Bobelyn, A. Peirs, W. Saeys, K. I. Theron, J. Lammertyn, Postharvest Biology and Technology, 46, N 2, 99–118 (2007).A. Wagner, S. Hilgert, T. Kattenborn, S. Fuchs, Water Supply, 19, N 4, 1204–1211 (2019).A. Wagner, S. Hilgert, T. Kattenborn, S. Fuchs, Water Supply, 19, N 4, 1204–1211 (2019).S. Katuwal, M. Knadel, T. Norgaard, P. Moldrup, M. H. Greve, L. W. de Jonge, Geoderma, 361, 114080 (2020).S. Katuwal, M. Knadel, T. Norgaard, P. Moldrup, M. H. Greve, L. W. de Jonge, Geoderma, 361, 114080 (2020).J.-H. Cheng, D.-W. Sun, Food Eng. Rev., 9, N 1, 36–49 (2017).J.-H. Cheng, D.-W. Sun, Food Eng. Rev., 9, N 1, 36–49 (2017).M. H. D. M. Ribeiro, L. dos Santos Coelho, Appl. Soft Computing, 86, 105837 (2020).M. H. D. M. Ribeiro, L. dos Santos Coelho, Appl. Soft Computing, 86, 105837 (2020).X. Luo, L. Xu, P. Huang, Y. Wang, J. Liu, Y. Hu, P. Wang, Z. Kang, Agriculture, 11, N 7, 673 (2021).X. Luo, L. Xu, P. Huang, Y. Wang, J. Liu, Y. Hu, P. Wang, Z. Kang, Agriculture, 11, N 7, 673 (2021).T. Chen, C. Guestrin, Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, 785–794 (2016).T. Chen, C. Guestrin, Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, 785–794 (2016).L. Prokhorenkova, G. Gusev, A. Vorobev, A. V. Dorogush, A. Gulin, arXiv preprint arXiv, 1706, 09516 (2017).L. Prokhorenkova, G. Gusev, A. Vorobev, A. V. Dorogush, A. Gulin, arXiv preprint arXiv, 1706, 09516 (2017).G. Ke, Q. Meng, T. Finley, T. Wang, W. Chen, W. Ma, Q. Ye, T.-Y. Liu, Adv. Neural Information Proc. Systems, 30, 3146–3154 (2017).G. Ke, Q. Meng, T. Finley, T. Wang, W. Chen, W. Ma, Q. Ye, T.-Y. Liu, Adv. Neural Information Proc. Systems, 30, 3146–3154 (2017).L. Breiman, Machine Learning, 45, N 1, 5–32 (2001).L. Breiman, Machine Learning, 45, N 1, 5–32 (2001).Y. Li, Y. Lei, P. Wang, M. Jiang, Y. Liu, Appl. Soft Computing, 101, 107003 (2021).Y. Li, Y. Lei, P. Wang, M. Jiang, Y. Liu, Appl. Soft Computing, 101, 107003 (2021).T. Wu, W. Zhang, X. Jiao, W. Guo, Y. Alhaj Hamoud, Computers and Electron. Agric., 184 (2021).T. Wu, W. Zhang, X. Jiao, W. Guo, Y. Alhaj Hamoud, Computers and Electron. Agric., 184 (2021).E. Al Daoud, Int. J. Computer and Information Engineering, 13, N 1, 6–10 (2019).E. Al Daoud, Int. J. Computer and Information Engineering, 13, N 1, 6–10 (2019).R. Zornoza, C. Guerrero, J. Mataix-Solera, K. Scow, V. Arcenegui, J. Mataix-Beneyto, Soil Biol. Biochem., 40, N 7, 1923–1930 (2008).R. Zornoza, C. Guerrero, J. Mataix-Solera, K. Scow, V. Arcenegui, J. Mataix-Beneyto, Soil Biol. Biochem., 40, N 7, 1923–1930 (2008).L. C. Lee, C.-Y. Liong, A. A. Jemain, AIP Conference Proceedings (AIP Publishing LLC, 020116 (2018).L. C. Lee, C.-Y. Liong, A. A. Jemain, AIP Conference Proceedings (AIP Publishing LLC, 020116 (2018).S. Chen, H. Xu, D. Xu, W. Ji, S. Li, M. Yang, B. Hu, Y. Zhou, N. Wang, D. Arrouays, Geoderma, 400, 115159 (2021).S. Chen, H. Xu, D. Xu, W. Ji, S. Li, M. Yang, B. Hu, Y. Zhou, N. Wang, D. Arrouays, Geoderma, 400, 115159 (2021).C. H. Bazoni, E. I. Ida, D. F. Barbin, L. E. Kurozawa, J. Stored Prod. Res., 73, 1–6 (2017).C. H. Bazoni, E. I. Ida, D. F. Barbin, L. E. Kurozawa, J. Stored Prod. Res., 73, 1–6 (2017).Y. Hong, S. Chen, Y. Liu, Y. Zhang, L. Yu, Y. Chen, Y. Liu, H. Cheng, Y. Liu, Catena, 174, 104–116 (2019).Y. Hong, S. Chen, Y. Liu, Y. Zhang, L. Yu, Y. Chen, Y. Liu, H. Cheng, Y. Liu, Catena, 174, 104–116 (2019).W. Ni, L. Nørgaard, M. Mørup, Anal. Chim. Acta, 813, 1–14 (2014).W. Ni, L. Nørgaard, M. Mørup, Anal. Chim. Acta, 813, 1–14 (2014).A. Kartakoullis, J. Comaposada, A. Cruz-Carrión, X. Serra, P. Gou, Food Chem., 278, 314–321 (2019).A. Kartakoullis, J. Comaposada, A. Cruz-Carrión, X. Serra, P. Gou, Food Chem., 278, 314–321 (2019).W. Ji, R. Viscarra Rossel, Z. Shi, Eur. J. Soil Sci., 66, N 3, 555–565 (2015).W. Ji, R. Viscarra Rossel, Z. Shi, Eur. J. Soil Sci., 66, N 3, 555–565 (2015).E. W. Ciurczak, Pract. Spectrosc. Ser., 27, 7–18 (2001).E. W. Ciurczak, Pract. Spectrosc. Ser., 27, 7–18 (2001).M. Haest, T. Cudahy, C. Laukamp, S. Gregory, Economic Geology, 107, N 2, 209–228 (2012).M. Haest, T. Cudahy, C. Laukamp, S. Gregory, Economic Geology, 107, N 2, 209–228 (2012).

The authors declare that there are no conflicts of interest present.