全球21.8%的疾病和22.7%的死亡歸因於環境因素，環境汙染防治已為全球的關鍵議題。我國自2000年訂定頒布相關法規，並於隔年開始限制輸入或產製在徵收範圍內各項物質的工業和企業，申報並繳納土壤及地下水汙染整治費。然根據現行的法規，製造物質係採自行申報制度，然此申報方式存在一定的風險，製造廠商可能透過不實申報以規避稅捐；另一方面，傳統的稽查過程耗費大量時間與人力成本，且政府每年例行的人工稽查對於逃漏稅捐的成效有限。文獻探討證實『使用數學算法可從現有數據找出可能的規避稅捐(詐騙)』。本研究借鑑金融產業已廣泛施行的機器學習詐騙偵測防阻技術，應用當代先進的集成式學習算法複合合成少數類過採樣技術(SMOTE)和自行開發的鑑識科學調查流程(Forensic-Based Investigation, FBI)元啟發式優化演算法，建構最佳化機器學習複合模型FBI-SMOTE-XGB，該模型在所有指標上均具有最卓越的表現，在Accuracy、Precision、Specificity、F1-Score和AUC績效指標分別達98.64%、97.98%、97.98%、98.64%和99.80%。FBI-SMOTE-XGB得自動化辨識並快速挑選重點稽查對象，進而向應該申報整治費卻未申報的製造廠商追討相應的欠稅，並將相關稅款回饋於解決土壤及地下水污染問題，改善生活環境並維護人民健康，為環境管理之永續發展做出重要貢獻。除此之外，數值實驗分析證明FBI元啟發式優化演算法在超參數調教方面具顯著的優勢，亦顯示不同模型經演算法最佳化調適後於績效指標的改善表現略有差異，因此研究證實超參數調教的必要性，應對所有的模型進行最佳化調適，以獲得在解決特定問題上最適用的模型。

21.8% of global diseases and 22.7% of deaths are attributed to environmental factors, making environmental pollution prevention and control a critical global issue. Since 2000, our country has established and promulgated relevant regulations. Starting the following year, it began to restrict the importation or production of various substances by industrial and commercial enterprises within the scope of collection, requiring declaration and payment of soil and groundwater pollution control fees. However, under the current regulations, the self-declaration system is adopted for manufacturing substances, which carries certain risks as manufacturers may exploit this method to evade taxes and duties through false reporting. Additionally, the traditional inspection process is time-consuming, costly in terms of manpower, and the government's routine manual inspections conducted annually have limited effectiveness in detecting tax evasion and underpayment. Literature studies have confirmed that "using mathematical algorithms to identify potential tax evasion (fraud) from existing data" is possible. This study draws on the widely implemented machine learning fraud detection and prevention techniques in the financial industry. It applies contemporary advanced integrated learning algorithms, such as the Synthesized Minority Oversampling Technique (SMOTE), and a self-developed Forensic-Based Investigation (FBI) algorithm. These components are used to construct an optimized machine learning hybrid model called FBI-SMOTE-XGB. The model demonstrates outstanding performance across all indicators, achieving a high accuracy of 98.64%, precision of 97.98%, specificity of 97.98%, F1-Score of 98.64%, and an AUC (Area Under the Curve) of 99.80%. FBI-SMOTE-XGB can automatically identify and quickly select key inspection targets, thereby pursuing corresponding unpaid taxes from manufacturing companies that should have declared but failed to do so. This contributes to the remediation of soil and groundwater pollution, improving living environments, safeguarding public health, and making important contributions to the sustainable development of environmental management. In addition, numerical experimental analysis has demonstrated the significant advantages of the FBI meta-heuristic optimization algorithm in hyperparameter tuning. It also indicates that different models show slight differences in performance improvement after algorithm optimization, highlighting the necessity of hyperparameter tuning. Therefore, all models should undergo optimization to obtain the most suitable model for addressing specific problems.

[1] H. Xu, Y. Jia, Z. Sun, J. Su, Q. S. Liu, Q. Zhou, and G. Jiang, "Environmental pollution, a hidden culprit for health issues," Eco-Environment & Health, vol. 1, no. 1, pp. 31-45, 2022/03/01/ 2022, doi: https://doi.org/10.1016/j.eehl.2022.04.003.
[2] A. Prüss-Üstün, J. Wolf, C. Corvalán, R. Bos, and M. Neira, "Preventing disease through healthy environments: a global assessment of the burden of disease from environmental risks," 2016, doi: https://www.who.int/publications/i/item/9789241565196.
[3] C. Han and X.-L. Tian, "Less pollution under a more centralized environmental system: Evidence from vertical environmental reforms in China," Energy Economics, vol. 112, p. 106121, 2022/08/01/ 2022, doi: https://doi.org/10.1016/j.eneco.2022.106121.
[4] 行政院環境保護署, "土壤及地下水污染整治法," 2000, doi: https://law.moj.gov.tw/LawClass/LawAll.aspx?pcode=O0110001.
[5] 行政院環境保護署, "土壤及地下水污染整治費收費辦法," 2001, doi: https://law.moj.gov.tw/LawClass/LawAll.aspx?pcode=O0110004.
[6] P. Gautam, S. Kumar, and S. Lokhandwala, "Advanced oxidation processes for treatment of leachate from hazardous waste landfill: A critical review," Journal of Cleaner Production, vol. 237, p. 117639, 2019/11/10/ 2019, doi: https://doi.org/10.1016/j.jclepro.2019.117639.
[7] W. Ahmad, R. D. Alharthy, M. Zubair, M. Ahmed, A. Hameed, and S. Rafique, "Toxic and heavy metals contamination assessment in soil and water to evaluate human health risk," Scientific Reports, vol. 11, no. 1, p. 17006, 2021/08/20 2021, doi: https://doi.org/10.1038/s41598-021-94616-4.
[8] O. O. Afolabi, E. Wali, E. C. Ihunda, M. C. Orji, V. O. Emelu, L. C. Bosco-Abiahu, N. C. Ogbuehi, S. O. Asomaku, and O. A. Wali, "Potential environmental pollution and human health risk assessment due to leachate contamination of groundwater from anthropogenic impacted site," Environmental Challenges, vol. 9, p. 100627, 2022/12/01/ 2022, doi: https://doi.org/10.1016/j.envc.2022.100627.
[9] S. Rezapour, A. Samadi, I. K. Kalavrouziotis, and N. Ghaemian, "Impact of the uncontrolled leakage of leachate from a municipal solid waste landfill on soil in a cultivated-calcareous environment," Waste Management, vol. 82, pp. 51-61, 2018/12/01/ 2018, doi: https://doi.org/10.1016/j.wasman.2018.10.013.
[10] H. Deng, Y. Tu, H. Wang, Z. Wang, Y. Li, L. Chai, W. Zhang, and Z. Lin, "Environmental behavior, human health effect, and pollution control of heavy metal(loid)s toward full life cycle processes," Eco-Environment & Health, vol. 1, no. 4, pp. 229-243, 2022/12/01/ 2022, doi: https://doi.org/10.1016/j.eehl.2022.11.003.
[11] S. Lai, J. Wu, Z. Ma, C. Ye, and H. Zhou, "Consumer Fraud Detection via P-feature Conversion," in 2021 IEEE 45th Annual Computers, Software, and Applications Conference (COMPSAC), 12-16 July 2021 2021, pp. 318-323, doi: https://doi.org/10.1109/COMPSAC51774.2021.00052.
[12] S. Ghosh and D. L. Reilly, "Credit card fraud detection with a neural-network," in System Sciences, 1994. Proceedings of the Twenty-Seventh Hawaii International Conference on, 1994, vol. 3: IEEE, pp. 621-630, doi: https://10.1109/HICSS.1994.323314.
[13] R. Van Belle, B. Baesens, and J. De Weerdt, "CATCHM: A novel network-based credit card fraud detection method using node representation learning," Decision Support Systems, vol. 164, p. 113866, 2023/01/01/ 2023, doi: https://doi.org/10.1016/j.dss.2022.113866.
[14] A. Rb and S. K. Kr, "Credit card fraud detection using artificial neural network," Global Transitions Proceedings, vol. 2, no. 1, pp. 35-41, 2021/06/01/ 2021, doi: https://doi.org/10.1016/j.gltp.2021.01.006.
[15] F. Aslam, A. I. Hunjra, Z. Ftiti, W. Louhichi, and T. Shams, "Insurance fraud detection: Evidence from artificial intelligence and machine learning," Research in International Business and Finance, vol. 62, p. 101744, 2022/12/01/ 2022, doi: https://doi.org/10.1016/j.ribaf.2022.101744.
[16] L. Settipalli, G. R. Gangadharan, and U. Fiore, "Predictive and adaptive Drift Analysis on Decomposed Healthcare Claims using ART based Topological Clustering," Information Processing & Management, vol. 59, no. 3, p. 102887, 2022/05/01/ 2022, doi: https://doi.org/10.1016/j.ipm.2022.102887.
[17] Y. Jiang, G. Liu, J. Wu, and H. Lin, "Telecom Fraud Detection via Hawkes-Enhanced Sequence Model," IEEE Transactions on Knowledge and Data Engineering, vol. 35, no. 5, pp. 5311-5324, 2023, doi: https://doi.org/10.1109/TKDE.2022.3150803.
[18] M. Habibpour, H. Gharoun, M. Mehdipour, A. Tajally, H. Asgharnezhad, A. Shamsi, A. Khosravi, and S. Nahavandi, "Uncertainty-aware credit card fraud detection using deep learning," Engineering Applications of Artificial Intelligence, vol. 123, p. 106248, 2023/08/01/ 2023, doi: https://doi.org/10.1016/j.engappai.2023.106248.
[19] H. Fanai and H. Abbasimehr, "A novel combined approach based on deep Autoencoder and deep classifiers for credit card fraud detection," Expert Systems with Applications, vol. 217, p. 119562, 2023/05/01/ 2023, doi: https://doi.org/10.1016/j.eswa.2023.119562.
[20] V. F. Rodrigues, L. M. Policarpo, D. E. da Silveira, R. da Rosa Righi, C. A. da Costa, J. L. V. Barbosa, R. S. Antunes, R. Scorsatto, and T. Arcot, "Fraud detection and prevention in e-commerce: A systematic literature review," Electronic Commerce Research and Applications, vol. 56, p. 101207, 2022/11/01/ 2022, doi: https://doi.org/10.1016/j.elerap.2022.101207.
[21] T. Shahana, V. Lavanya, and A. R. Bhat, "State of the art in financial statement fraud detection: A systematic review," Technological Forecasting and Social Change, vol. 192, p. 122527, 2023/07/01/ 2023, doi: https://doi.org/10.1016/j.techfore.2023.122527.
[22] Y. Gao, X. Yin, Z. He, and X. Wang, "A deep learning process anomaly detection approach with representative latent features for low discriminative and insufficient abnormal data," Computers & Industrial Engineering, vol. 176, p. 108936, 2023/02/01/ 2023, doi: https://doi.org/10.1016/j.cie.2022.108936.
[23] A. T. Khan, X. Cao, S. Li, V. N. Katsikis, I. Brajevic, and P. S. Stanimirovic, "Fraud detection in publicly traded U.S firms using Beetle Antennae Search: A machine learning approach," Expert Systems with Applications, vol. 191, p. 116148, 2022/04/01/ 2022, doi: https://doi.org/10.1016/j.eswa.2021.116148.
[24] A. Correa Bahnsen, D. Aouada, A. Stojanovic, and B. Ottersten, "Feature engineering strategies for credit card fraud detection," Expert Systems with Applications, vol. 51, pp. 134-142, 2016/06/01/ 2016, doi: https://doi.org/10.1016/j.eswa.2015.12.030.
[25] Q. C. Tran, D. D. Minh, A. Jaafari, N. Al-Ansari, D. D. Minh, D. T. Van, D. A. Nguyen, T. H. Tran, L. S. Ho, D. H. Nguyen, I. Prakash, H. V. Le, and B. T. Pham, "Novel Ensemble Landslide Predictive Models Based on the Hyperpipes Algorithm: A Case Study in the Nam Dam Commune, Vietnam," Applied Sciences, vol. 10, no. 11, doi: https://doi.org/10.3390/app10113710.
[26] B. T. Pham, T. Nguyen-Thoi, C. Qi, T. V. Phong, J. Dou, L. S. Ho, H. V. Le, and I. Prakash, "Coupling RBF neural network with ensemble learning techniques for landslide susceptibility mapping," CATENA, vol. 195, p. 104805, 2020/12/01/ 2020, doi: https://doi.org/10.1016/j.catena.2020.104805.
[27] H. Dong, D. He, and F. Wang, "SMOTE-XGBoost using Tree Parzen Estimator optimization for copper flotation method classification," Powder Technology, vol. 375, pp. 174-181, 2020/09/20/ 2020, doi: https://doi.org/10.1016/j.powtec.2020.07.065.
[28] A. Shakeel, D. Chong, and J. Wang, "District heating load forecasting with a hybrid model based on LightGBM and FB-prophet," Journal of Cleaner Production, vol. 409, p. 137130, 2023/07/10/ 2023, doi: https://doi.org/10.1016/j.jclepro.2023.137130.
[29] Z. Yan, H. Chen, X. Dong, K. Zhou, and Z. Xu, "Research on prediction of multi-class theft crimes by an optimized decomposition and fusion method based on XGBoost," Expert Systems with Applications, vol. 207, p. 117943, 2022/11/30/ 2022, doi: https://doi.org/10.1016/j.eswa.2022.117943.
[30] Z. Lao, D. He, Z. Wei, H. Shang, Z. Jin, J. Miao, and C. Ren, "Intelligent fault diagnosis for rail transit switch machine based on adaptive feature selection and improved LightGBM," Engineering Failure Analysis, vol. 148, p. 107219, 2023/06/01/ 2023, doi: https://doi.org/10.1016/j.engfailanal.2023.107219.
[31] G. Mohiuddin, Z. Lin, J. Zheng, J. Wu, W. Li, Y. Fang, S. Wang, J. Chen, and X. Zeng, "Intrusion detection using hybridized meta-heuristic techniques with Weighted XGBoost Classifier," Expert Systems with Applications, p. 120596, 2023/06/05/ 2023, doi: https://doi.org/10.1016/j.eswa.2023.120596.
[32] C.-H. Cheng, Y.-F. Kao, and H.-P. Lin, "A financial statement fraud model based on synthesized attribute selection and a dataset with missing values and imbalanced classes," Applied Soft Computing, vol. 108, p. 107487, 2021/09/01/ 2021, doi: https://doi.org/10.1016/j.asoc.2021.107487.
[33] S. H. Khan, M. Hayat, M. Bennamoun, F. A. Sohel, and R. Togneri, "Cost-Sensitive Learning of Deep Feature Representations From Imbalanced Data," IEEE Transactions on Neural Networks and Learning Systems, vol. 29, no. 8, pp. 3573-3587, 2018, doi: https://doi.org/10.1109/TNNLS.2017.2732482.
[34] N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, "SMOTE: synthetic minority over-sampling technique," Journal of artificial intelligence research, vol. 16, pp. 321-357, 2002, doi: https://doi.org/10.48550/arXiv.1106.1813.
[35] Z. Xu, D. Shen, T. Nie, and Y. Kou, "A hybrid sampling algorithm combining M-SMOTE and ENN based on Random forest for medical imbalanced data," Journal of Biomedical Informatics, vol. 107, p. 103465, 2020/07/01/ 2020, doi: https://doi.org/10.1016/j.jbi.2020.103465.
[36] A. Li, M. Liu, and S. Sheather, "Predicting stock splits using ensemble machine learning and SMOTE oversampling," Pacific-Basin Finance Journal, vol. 78, p. 101948, 2023/04/01/ 2023, doi: https://doi.org/10.1016/j.pacfin.2023.101948.
[37] Y. Liu, Z. Yu, C. Chen, Y. Han, and B. Yu, "Prediction of protein crotonylation sites through LightGBM classifier based on SMOTE and elastic net," Analytical Biochemistry, vol. 609, p. 113903, 2020/11/15/ 2020, doi: https://doi.org/10.1016/j.ab.2020.113903.
[38] L. Han, G. Yang, X. Yang, X. Song, B. Xu, Z. Li, J. Wu, H. Yang, and J. Wu, "An explainable XGBoost model improved by SMOTE-ENN technique for maize lodging detection based on multi-source unmanned aerial vehicle images," Computers and Electronics in Agriculture, vol. 194, p. 106804, 2022/03/01/ 2022, doi: https://doi.org/10.1016/j.compag.2022.106804.
[39] X. Hao, Z. Zhang, Q. Xu, G. Huang, and K. Wang, "Prediction of f-CaO content in cement clinker: A novel prediction method based on LightGBM and Bayesian optimization," Chemometrics and Intelligent Laboratory Systems, vol. 220, p. 104461, 2022/01/15/ 2022, doi: https://doi.org/10.1016/j.chemolab.2021.104461.
[40] B. Ibrahim, I. Ahenkorah, A. Ewusi, and F. Majeed, "A novel XRF-based lithological classification in the Tarkwaian paleo placer formation using SMOTE-XGBoost," Journal of Geochemical Exploration, vol. 245, p. 107147, 2023/02/01/ 2023, doi: https://doi.org/10.1016/j.gexplo.2022.107147.
[41] K. Jas and G. R. Dodagoudar, "Explainable machine learning model for liquefaction potential assessment of soils using XGBoost-SHAP," Soil Dynamics and Earthquake Engineering, vol. 165, p. 107662, 2023/02/01/ 2023, doi: https://doi.org/10.1016/j.soildyn.2022.107662.
[42] H. Yu-Chi and D. L. Pepyne, "Simple explanation of the no free lunch theorem of optimization," in Proceedings of the 40th IEEE Conference on Decision and Control (Cat. No.01CH37228), 4-7 Dec. 2001 2001, vol. 5, pp. 4409-4414 vol.5, doi: https://doi.org/10.1109/CDC.2001.980896.
[43] J. Bertomeu, E. Cheynel, E. Floyd, and W. Pan, "Using machine learning to detect misstatements," Review of Accounting Studies, vol. 26, no. 2, pp. 468-519, 2021/06/01 2021, doi: https://doi.org/10.1007/s11142-020-09563-8.
[44] Y. Xia, C. Liu, Y. Li, and N. Liu, "A boosted decision tree approach using Bayesian hyper-parameter optimization for credit scoring," Expert Systems with Applications, vol. 78, pp. 225-241, 2017/07/15/ 2017, doi: https://doi.org/10.1016/j.eswa.2017.02.017.
[45] A. Buabeng, A. Simons, N. K. Frempong, and Y. Y. Ziggah, "A novel hybrid predictive maintenance model based on clustering, smote and multi-layer perceptron neural network optimised with grey wolf algorithm," SN Applied Sciences, vol. 3, no. 5, p. 593, 2021/04/30 2021, doi: https://doi.org/10.1007/s42452-021-04598-1.
[46] Q. Pan, C. Zhang, X. Wei, A. Wan, and Z. Wei, "Assessment of MV XLPE cable aging state based on PSO-XGBoost algorithm," Electric Power Systems Research, vol. 221, p. 109427, 2023/08/01/ 2023, doi: https://doi.org/10.1016/j.epsr.2023.109427.
[47] S. Gao and C. W. de Silva, "Estimation distribution algorithms on constrained optimization problems," Applied Mathematics and Computation, vol. 339, pp. 323-345, 2018/12/15/ 2018, doi: https://doi.org/10.1016/j.amc.2018.07.037.
[48] T. Mao, A. S. Mihăită, F. Chen, and H. L. Vu, "Boosted Genetic Algorithm Using Machine Learning for Traffic Control Optimization," IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 7, pp. 7112-7141, 2022, doi: https://doi.org/10.1109/TITS.2021.3066958.
[49] M. M. Rosso, R. Cucuzza, F. Di Trapani, and G. C. Marano, "Nonpenalty Machine Learning Constraint Handling Using PSO-SVM for Structural Optimization," Advances in Civil Engineering, vol. 2021, p. 6617750, 2021/02/16 2021, doi: https://doi.org/10.1155/2021/6617750.
[50] J.-S. Chou and N.-M. Nguyen, "FBI inspired meta-optimization," Applied Soft Computing, vol. 93, p. 106339, 2020/08/01/ 2020, doi: https://doi.org/10.1016/j.asoc.2020.106339.
[51] D. H. D. Tran, H. Q.-P. D. Le, N.-T. Nguyen, and T.-T. D. Le, "Robust forensic-based investigation algorithm for resource leveling in multiple projects," Scientia Iranica, 2022, doi: https://doi.org/10.24200/SCI.2022.60318.6731.
[52] J.-S. Chou, C.-Y. Liu, H. Prayogo, R. R. Khasani, D. Gho, and G. G. Lalitan, "Predicting nominal shear capacity of reinforced concrete wall in building by metaheuristics-optimized machine learning," Journal of Building Engineering, vol. 61, p. 105046, 2022/12/01/ 2022, doi: https://doi.org/10.1016/j.jobe.2022.105046.
[53] J.-S. Chou, L.-Y. Chen, and C.-Y. Liu, "Forensic-based investigation-optimized extreme gradient boosting system for predicting compressive strength of ready-mixed concrete," Journal of Computational Design and Engineering, vol. 10, no. 1, pp. 425-445, 2023, doi: https://doi.org/10.1093/jcde/qwac133.
[54] A. Fathy, H. Rezk, and T. M. Alanazi, "Recent Approach of Forensic-Based Investigation Algorithm for Optimizing Fractional Order PID-Based MPPT With Proton Exchange Membrane Fuel Cell," IEEE Access, vol. 9, pp. 18974-18992, 2021, doi: https://doi.org/10.1109/ACCESS.2021.3054552.
[55] J. Sahu, P. Satapathy, B. K. Sahu, P. K. Mohanty, and A. K. Naik, "Design of a quasi-oppositional FBI based fuzzy PIDF (1 + PI) controller for frequency regulation with renewable energy sources and energy storage elements," International Journal of Numerical Modelling: Electronic Networks, Devices and Fields, https://doi.org/10.1002/jnm.3120 vol. n/a, no. n/a, p. e3120, 2023/05/08 2023, doi: https://doi.org/10.1002/jnm.3120.
[56] M.-T. Cao, N.-M. Nguyen, K.-T. Chang, X.-L. Tran, and N.-D. Hoang, "Automatic recognition of concrete spall using image processing and metaheuristic optimized LogitBoost classification tree," Advances in Engineering Software, vol. 159, p. 103031, 2021/09/01/ 2021, doi: https://doi.org/10.1016/j.advengsoft.2021.103031.
[57] M.-T. Cao, K.-T. Chang, N.-M. Nguyen, V.-D. Tran, X.-L. Tran, and N.-D. Hoang, "Image processing-based automatic detection of asphalt pavement rutting using a novel metaheuristic optimized machine learning approach," Soft Computing, vol. 25, no. 20, pp. 12839-12855, 2021/10/01 2021, doi: https://doi.org/10.1007/s00500-021-06086-5.
[58] Y. Tang, K. Blincoe, and A. W. Kempa-Liehr, "Enriching feature engineering for short text samples by language time series analysis," EPJ Data Science, vol. 9, no. 1, p. 26, 2020/08/31 2020, doi: https://doi.org/10.1140/epjds/s13688-020-00244-9.
[59] W. McKinney, "pandas," 2008, doi: https://github.com/pandas-dev/pandas.
[60] D. Cournapeau, "scikit-learn," 2007, doi: https://github.com/scikit-learn/scikit-learn.
[61] H. Tin Kam, "Random decision forests," in Proceedings of 3rd International Conference on Document Analysis and Recognition, 14-16 Aug. 1995 1995, vol. 1, pp. 278-282 vol.1, doi: https://doi.org/10.1109/ICDAR.1995.598994.
[62] L. Breiman, "Random Forests," Machine Learning, vol. 45, no. 1, pp. 5-32, 2001/10/01 2001, doi: https://doi.org/10.1023/A:1010933404324.
[63] G. Biau and E. Scornet, "A random forest guided tour," TEST, vol. 25, no. 2, pp. 197-227, 2016/06/01 2016, doi: https://doi.org/10.1007/s11749-016-0481-7.
[64] G. Ke, Q. Meng, T. Finley, T. Wang, W. Chen, W. Ma, Q. Ye, and T.-Y. Liu, "Lightgbm: A highly efficient gradient boosting decision tree," Advances in neural information processing systems, vol. 30, 2017, doi: https://proceedings.neurips.cc/paper/2017/hash/6449f44a102fde848669bdd9eb6b76fa-Abstract.html.
[65] T. Chen and C. Guestrin, "Xgboost: A scalable tree boosting system," in Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, 2016, pp. 785-794, doi: https://doi.org/10.48550/arXiv.1603.02754.
[66] S. W. Greenhouse and S. Geisser, "On methods in the analysis of profile data," Psychometrika, vol. 24, no. 2, pp. 95-112, 1959/06/01 1959, doi: https://doi.org/10.1007/BF02289823.
[67] J.-S. Chou, C.-F. Tsai, A.-D. Pham, and Y.-H. Lu, "Machine learning in concrete strength simulations: Multi-nation data analytics," Construction and Building Materials, vol. 73, pp. 771-780, 2014/12/30/ 2014, doi: https://doi.org/10.1016/j.conbuildmat.2014.09.054.
[68] R. Kohavi, "A study of cross-validation and bootstrap for accuracy estimation and model selection," in Ijcai, 1995, vol. 14, no. 2: Montreal, Canada, pp. 1137-1145, doi: https://dl.acm.org/doi/10.5555/1643031.1643047.