Predictive Modeling of Korean Traditional Paper Characteristics Using Machine Learning Approaches (Part 2): Prediction of Carbonyl Content and Analysis of Variable Importance Using Random Forest

Kang-Jae Kim; Jin-Ho Kim; Geunyong Park; Myung-Joon Jeong

doi:10.7584/JKTAPPI.2023.10.55.5.13

All Issue

2023 Vol.55, Issue 5 Preview Page Next Page

Research Article

Predictive Modeling of Korean Traditional Paper Characteristics Using Machine Learning Approaches (Part 2): Prediction of Carbonyl Content and Analysis of Variable Importance Using Random Forest 기계학습을 이용한 한지의 특성 예측 모델링 (Part 2) -랜덤 포레스트를 이용한 카보닐기 함량 예측과 변수 중요도 분석-

30 October 2023. pp. 13-23

PDF XML

Abstract

This paper introduces a random forest regression model trained with infrared spectral data to predict the carbonyl content of Hanji, a traditional Korean paper. The random forest model demonstrated excellent performance in carbonyl content prediction, surpassing the results obtained from the partial least squares model. To optimize the infrared spectral range for prediction, the spectral range was restricted from the entire range of 4000-400 cm^-1 to the narrower range of 1800-1200 cm^-1, known for its suitability in characterizing paper properties. This limitation enhanced the coefficients of determination of the model, increasing it from 0.921 to 0.937. A permutation variable importance measure was then applied to identify the key spectral regions contributing to carbonyl content prediction. The analysis pinpointed the 1650-1350 cm^-1 range as a crucial region for accurate predictions. Subsequently, a new prediction model was built using data exclusively from this important region, yielding remarkably improved coefficients of determination of 0.960 and 0.965 for the raw and second derivative spectra, respectively. These findings affirm the validity and significance of the critical region identified by the permutation variable importance measure. The predictive performance of the established models is valid within the range of 7.2 to 29.4 μmol/g of carbonyl content in the training set.

Keywords

Carbonyl content

Hanji, infrared spectroscopy

machine learning

permutation importance

random forest

regression

References

Jeong, S. H., A study on manufacturing technologies and excellence of Korean traditional paper, Korean Journal of Cultural Heritage Studies 48(1):96-131 (2015).

Han, B., Vilmont, L. B., Kim, H. J., Lavédrine, B., Sakamoto, S., and Sablier, M., Characterization of Korean handmade papers collected in a Hanji reference book, Heritage Science 9:1-12 (2021). 10.1186/s40494-021-00570-9

Kim, K. J., and Eom, T. J., Study on the aging characteristics of paper with principal component analysis, Journal of Korea TAPPI 48(6):144-149 (2016). 10.7584/JKTAPPI.2016.12.48.6.144

Brown, S. D., Blank, T. B., Sum, S. T., and Weyer, L. G., Chemometrics, Analytical Chemistry 66(12):315-359 (1994). 10.1021/ac00084a0148092471

Hwang, S. W., Horikawa, Y., Lee, W. H., and Sugiyama, J., Identification of Pinus species related to historic architecture in Korea using NIR chemometric approaches, Journal of Wood Science 62(2):156-167 (2016). 10.1007/s10086-016-1540-0

Hwang, S. W., Hwang, U. T., Jo, K., Lee, T., Park, J., Kim, J. C., Kwak, H. W., Choi, I. G., and Yeo, H., NIR-chemometric approaches for evaluating carbonization characteristics of hydrothermally carbonized lignin, Scientific Reports 11(1):16979 (2021). 10.1038/s41598-021-96461-x34417504PMC8379198

Kim, K. J., Ahn, E. B., Ryu, J. A., and Eom, T. J., Identification of pulp fibers in the chronological academic journals using spectroscopic method, Journal of Korea TAPPI 50(4):61-70 (2018). 10.7584/JKTAPPI.2018.08.50.4.61

Oliveri, P., Malegori, C., and Casale, M., Chemometrics: Multivariate analysis of chemical data. Chemical Analysis of Food, Academic Press (2020). 10.1016/B978-0-12-813266-1.00002-4

Meza Ramirez, C. A., Greenop, M., Ashton, L., and Rehman, I. U., Applications of machine learning in spectroscopy, Applied Spectroscopy Reviews 56(8-10): 733-763 (2021). 10.1080/05704928.2020.1859525

Fu, W., and Hopkins, W. S., Applying machine learning to vibrational spectroscopy, The Journal of Physical Chemistry A 122(1):167-171 (2018). 10.1021/acs.jpca.7b1030329211476

Breiman, L., Random forests, Machine Learning 45:5-32 (2001). 10.1023/A:1010933404324

Heo, T. I., Kim, D. H., and Hwang, S. W., Identification of Celtis species using random forest with infrared spectroscopy and analysis of spectral feature importance, The Korean Data & Information Science Society 32(6):1183-1194 (2021). 10.7465/jkdi.2021.32.6.1183

Hwang, S. W., Chung, H., Lee, T., Kim, J., Kim, Y., Kim, J. C., Kwak, H. W., Choi, I. G., and Yeo, H., Feature importance measures from random forest regressor using near-infrared spectra for predicting carbonization characteristics of kraft lignin-derived hydrochar, Journal of Wood Science 69:1 (2023). 10.1186/s10086-022-02073-y

Cutler, D. R., Edwards Jr, T. C., Beard, K. H., Cutler, A., Hess, K. T., Gibson, J., and Lawler, J. J., Random forests for classification in ecology, Ecology 88(11):2783-2792 (2007). 10.1890/07-0539.118051647

Hwang, S. W., Park, G., Kim, J., Jeong, M. J., Predictive modeling of traditional Korean paper characteristics using machine learning approaches (part 1): Discriminating manufacturing origins with artificial neural networks and infrared spectroscopy, Journal of Korea TAPPI 55(4):57-69 (2023). 10.7584/JKTAPPI.2023.8.55.4.57

Jeong, M. J., Kang, K. Y., Bacher, M., Kim, H. J., Jo, B. M., and Potthast, A., Deterioration of ancient cellulose paper, Hanji: evaluation of paper permanence, Cellulose 21:4621-4632 (2014). 10.1007/s10570-014-0455-4

Savitzky, A., and Golay, M. J., Smoothing and differentiation of data by simplified least squares procedures, Analytical Chemistry 36(8):1627-1639 (1964). 10.1021/ac60214a047

Breiman, L., Friedman, J., Stone, C. J., and Olshen, R. A., Classification and regression trees, CRC press, (1984).

Hwang, S. W., Chung, H., Lee, T., Kwak, H. W., Choi, I. G., and Yeo, H., Investigation of NIR spectroscopy and electrical resistance-based approaches for moisture determination of logging residues and sweet sorghum, BioResources 18(1):2064-2028 (2023). 10.15376/biores.18.1.2064-2082

Maréchal, Y., and Chanzy, H., The hydrogen bond network in Iβ cellulose as observed by infrared spectrometry, Journal of Molecular Structure 523(1-3):183-196 (2000). 10.1016/S0022-2860(99)00389-0

Xiao, S., Gao, R., Lu, Y., Li, J., and Sun, Q., Fabrication and characterization of nanofibrillated cellulose and its aerogels from natural pine needles, Carbohydrate Polymers 119:202-209 (2015). 10.1016/j.carbpol.2014.11.04125563961

Garside, P., and Wyeth, P., Identification of cellulosic fibres by FTIR spectroscopy-thread and single fibre analysis by attenuated total reflectance, Studies in Conservation 48(4):269-275 (2003). 10.1179/sic.2003.48.4.269

Pandey, K. K., A study of chemical structure of soft and hardwood and wood polymers by FTIR spectroscopy, Journal of Applied Polymer Science 71(12):1969-1975 (1999). 10.1002/(SICI)1097-4628(19990321)71:12<1969::AID-APP6>3.0.CO;2-D

Ciolacu, D., Kovac, J., and Kokol, V., The effect of the cellulose-binding domain from Clostridium cellulovorans on the supramolecular structure of cellulose fibers, Carbohydrate Research 345(5):621-630 (2010). 10.1016/j.carres.2009.12.02320122684

Castro, K., Princi, E., Proietti, N., Manso, M., Capitani, D., Vicini, S., Madarigana, J. M., and De Carvalho, M. L., Assessment of the weathering effects on cellulose based materials through a multianalytical approach, Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 269(12):1401-1410 (2011). 10.1016/j.nimb.2011.03.027

Delmotte, L., Ganne-Chedeville, C., Leban, J. M., Pizzi, A., and Pichelin, F., CP-MAS 13C NMR and FT-IR investigation of the degradation reactions of polymer constituents in wood welding, Polymer Degradation and Stability 93(2):406-412 (2008). 10.1016/j.polymdegradstab.2007.11.020

Lee, J. M., Ahn, E. B., Choi, H. S., Ryu, J. A., and Eom, T. J., Study on improvement of preparation efficiency of lignin rich micro-cellulosic fines, Journal of Korea TAPPI 50(5):114-122 (2018). 10.7584/JKTAPPI.2018.10.50.5.114

Polovka, M., Polovková, J., Vizárová, K., Kirschnerová, S., Bieliková, L., and Vrška, M., The application of FTIR spectroscopy on characterization of paper samples, modified by Bookkeeper process, Vibrational Spectroscopy 41(1):112-117 (2006). 10.1016/j.vibspec.2006.01.010

Lee, Y., Ryu, J. A., Kim, K. T., and Kim, K. J., Scientific approach to confirm the excellence of Seokgayeoraehaengjeoksong as Korean cultural heritage, Journal of Korea TAPPI 52(6):73-81 (2020). 10.7584/JKTAPPI.2020.12.52.6.73

Dupont, A. L., Degradation of cellulose at the wet/dry interface II - An approach to the identification of the oxidation compounds, Restaurator 17:145-164 (1996). 10.1515/rest.1996.17.3.145

Information

Publisher :Korea Technical Association of The Pulp and Paper Industry
Publisher(Ko) :한국펄프종이공학회
Journal Title :Journal of Korea TAPPI
Journal Title(Ko) :펄프종이기술
Volume : 55
No :5
Pages :13-23
Received Date : 2023-07-27
Revised Date : 2023-10-19
Accepted Date : 2023-10-20
DOI :https://doi.org/10.7584/JKTAPPI.2023.10.55.5.13

Journal of Korea TAPPI ISSN:0253-3200(Print) 펄프종이기술

All Issue