{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,9]],"date-time":"2026-07-09T03:47:04Z","timestamp":1783568824505,"version":"3.55.0"},"reference-count":30,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2023,1,20]],"date-time":"2023-01-20T00:00:00Z","timestamp":1674172800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Digital agriculture employs artificial intelligence (AI) to transform data collected in the field into actionable crop management. Effective digital agriculture models can detect problems early, reducing costs significantly. However, ineffective models can be counterproductive. Farmers often want to validate models by spot checking their fields before expending time and effort on recommended actions. However, in large fields, farmers can spot check too few areas, leading them to wrongly believe that ineffective models are effective. Model validation is especially difficult for models that use neural networks, an AI technology that normally assesses crops health accurately but makes inexplicable recommendations. We present a new approach that trains random forests, an AI modeling approach whose recommendations are easier to explain, to mimic neural network models. Then, using the random forest as an explainable white box, we can (1) gain knowledge about the neural network, (2) assess how well a test set represents possible inputs in a given field, (3) determine when and where a farmer should spot check their field for model validation, and (4) find input data that improve the test set. We tested our approach with data used to assess soybean defoliation. Using information from the four processes above, our approach can reduce spot checks by up to 94%.<\/jats:p>","DOI":"10.3390\/s23031187","type":"journal-article","created":{"date-parts":[[2023,1,20]],"date-time":"2023-01-20T02:35:55Z","timestamp":1674182155000},"page":"1187","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":19,"title":["AI-Driven Validation of Digital Agriculture Models"],"prefix":"10.3390","volume":"23","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1803-8176","authenticated-orcid":false,"given":"Eduardo","family":"Romero-Gainza","sequence":"first","affiliation":[{"name":"Department of Computer Science and Engineering, The Ohio State University, Columbus, OH 43210, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christopher","family":"Stewart","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, The Ohio State University, Columbus, OH 43210, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2023,1,20]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"105836","DOI":"10.1016\/j.compag.2020.105836","article-title":"Detection and classification of soybean pests using deep learning with UAV images","volume":"179","author":"Tetila","year":"2020","journal-title":"Comput. Electron. Agric."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"22","DOI":"10.1016\/j.compag.2017.05.001","article-title":"An overview of current and potential applications of thermal remote sensing in precision agriculture","volume":"139","author":"Khanal","year":"2017","journal-title":"Comput. Electron. Agric."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Zhang, Z., Boubin, J., Stewart, C., and Khanal, S. (2020). Whole-Field Reinforcement Learning: A Fully Autonomous Aerial Scouting Method for Precision Agriculture. Sensors, 20.","DOI":"10.3390\/s20226585"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Wang, S.C. (2003). Artificial neural network. Interdisciplinary Computing in Java Programming, Springer.","DOI":"10.1007\/978-1-4615-0377-4"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"106682","DOI":"10.1016\/j.compag.2021.106682","article-title":"Assessing the efficacy of machine learning techniques to characterize soybean defoliation from unmanned aerial vehicles","volume":"193","author":"Zhang","year":"2022","journal-title":"Comput. Electron. Agric."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"012039","DOI":"10.1088\/1742-6596\/1969\/1\/012039","article-title":"Deep learning algorithms for tool condition monitoring in milling: A review","volume":"1969","author":"Patil","year":"2021","journal-title":"J. Phys. Conf. Ser. IOP Publ."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"71","DOI":"10.1145\/234313.234346","article-title":"Learning decision tree classifiers","volume":"28","author":"Quinlan","year":"1996","journal-title":"ACM Comput. Surv. (CSUR)"},{"key":"ref_8","unstructured":"Deo, T.Y., Patange, A.D., Pardeshi, S.S., Jegadeeshwaran, R., Khairnar, A.N., and Khade, H.S. (2021). A white-box SVM framework and its swarm-based optimization for supervision of toothed milling cutter through characterization of spindle vibrations. arXiv."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1283","DOI":"10.1016\/j.matpr.2021.02.128","article-title":"Design of bagged tree ensemble for carbide coated inserts fault diagnosis","volume":"46","author":"Khade","year":"2021","journal-title":"Mater. Today Proc."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"229","DOI":"10.23940\/ijpe.21.02.p7.229240","article-title":"Supervision of Carbide Tool Condition by Training of Vibration-based Statistical Model using Boosted Trees Ensemble","volume":"17","author":"Khairnar","year":"2021","journal-title":"Int. J. Perform. Eng."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"012049","DOI":"10.1088\/1742-6596\/1969\/1\/012049","article-title":"Data Driven Cutting Tool Fault Diagnosis System Using Machine Learning Approach: A Review","volume":"1969","author":"Tambake","year":"2021","journal-title":"J. Phys. Conf. Ser. IOP Publ."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"621","DOI":"10.1239\/jap\/1222441818","article-title":"The Generalised Coupon Collector Problem","volume":"45","author":"Neal","year":"2008","journal-title":"J. Appl. Probab."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"1773","DOI":"10.1021\/acs.jcim.6b00753","article-title":"Comparison of the predictive performance and interpretability of random forest and linear models on benchmark data sets","volume":"57","author":"Palczewska","year":"2017","journal-title":"J. Chem. Inf. Model."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Romero-Gainza, E., Stewart, C., Li, A., Hale, K., and Morris, N. (2021, January 9\u201312). Memory Mapping and Parallelizing Random Forests for Speed and Cache Efficiency. Proceedings of the 50th International Conference on Parallel Processing Workshop, Lemont, IL, USA. ICPP Workshops \u201921.","DOI":"10.1145\/3458744.3474052"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Romero, E., Stewart, C., Li, A., Hale, K., and Morris, N. (2022, January 7\u201311). Bolt: Fast Inference for Random Forests. Proceedings of the 23rd ACM\/IFIP International Middleware Conference, Quebec, QC, Canada.","DOI":"10.1145\/3528535.3531519"},{"key":"ref_17","unstructured":"O\u2019Shea, K., and Nash, R. (2015). An introduction to convolutional neural networks. arXiv."},{"key":"ref_18","first-page":"2825","article-title":"Scikit-learn: Machine Learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"72","DOI":"10.1016\/S0958-1669(02)00288-4","article-title":"Predictive non-linear modeling of complex data by artificial neural networks","volume":"13","author":"Almeida","year":"2002","journal-title":"Curr. Opin. Biotechnol."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Lee, J., Xiao, L., Schoenholz, S., Bahri, Y., Novak, R., Sohl-Dickstein, J., and Pennington, J. (2019). Wide neural networks of any depth evolve as linear models under gradient descent. Advances in Neural Information Processing Systems, Curran Associates, Inc.","DOI":"10.1088\/1742-5468\/abc62b"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"1178","DOI":"10.1016\/j.ejor.2021.06.053","article-title":"Machine learning for credit scoring: Improving logistic regression with non-linear decision-tree effects","volume":"297","author":"Dumitrescu","year":"2022","journal-title":"Eur. J. Oper. Res."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"102899","DOI":"10.1016\/j.cities.2020.102899","article-title":"Examining non-linear associations between population density and waist-hip ratio: An application of gradient boosting decision trees","volume":"107","author":"Yin","year":"2020","journal-title":"Cities"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"183","DOI":"10.1016\/j.eswa.2018.12.041","article-title":"Inducing non-orthogonal and non-linear decision boundaries in decision trees via interactive basis functions","volume":"122","author":"Paez","year":"2019","journal-title":"Expert Syst. Appl."},{"key":"ref_24","first-page":"46","article-title":"Definite integrals, Riemann sums, and area under a curve: What is necessary and sufficient","volume":"Volume 2","author":"Sealey","year":"2006","journal-title":"Proceedings of the 28th Annual Meeting of the North American Chapter of the International Group for the Psychology of Mathematics Education"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"1999","DOI":"10.1016\/S0031-3203(98)00181-2","article-title":"Extracting decision trees from trained neural networks","volume":"32","author":"Krishnan","year":"1999","journal-title":"Pattern Recognit."},{"key":"ref_26","unstructured":"Craven, M.W., and Shavlik, J.W. (1995, January 27\u201330). Extracting Tree-Structured Representations of Trained Networks. Proceedings of the NIPS, Denver, CO, USA."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Johansson, U., and Niklasson, L. (April, January 30). Evolving Decision Trees Using Oracle Guides. Proceedings of the 2009 IEEE Symposium on Computational Intelligence and Data Mining, Nashville, TN, USA.","DOI":"10.1109\/CIDM.2009.4938655"},{"key":"ref_28","first-page":"26","article-title":"Please stop explaining black box models for high stakes decisions","volume":"1050","author":"Rudin","year":"2018","journal-title":"Stat"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1021\/ci0342472","article-title":"The problem of overfitting","volume":"44","author":"Hawkins","year":"2004","journal-title":"J. Chem. Inf. Comput. Sci."},{"key":"ref_30","unstructured":"Hong, S., You, T., Kwak, S., and Han, B. (2015, January 6\u201311). Online Tracking by Learning Discriminative Saliency Map with Convolutional Neural Network. Proceedings of the 32nd International Conference on International Conference on Machine Learning (ICML 2015), Lille, France. Available online: https:\/\/proceedings.mlr.press\/v37\/hong15.html."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/23\/3\/1187\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T18:11:53Z","timestamp":1760119913000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/23\/3\/1187"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,1,20]]},"references-count":30,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2023,2]]}},"alternative-id":["s23031187"],"URL":"https:\/\/doi.org\/10.3390\/s23031187","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,1,20]]}}}