Abstract
Machine Learning (ML) is becoming an increasingly critical technology in many areas such as health, business but also in everyday applications of significant societal importance. However, the lack of explainability or ability of ML systems to offer explanation on how they work, which refers to the model (related to the whole data) and sample explainability (related to specific samples) poses significant challenges in their adoption, verification, and in ensuring the trust among users and general public. We present novel integrated Random Forest Model and Sample Explainer – RFEX. RFEX is specifically designed for important class of users who are non-ML experts but are often the domain experts and key decision makers. RFEX provides easy to analyze one-page Model and Sample explainability summaries in tabular format with wealth of explainability information including classification confidence, tradeoff between accuracy and features used, as well as ability to identify potential outlier samples and features. We demonstrate RFEX on two case studies: mortality prediction for COVID-19 patients from the data obtained from Huazhong University of Science and Technology, Wuhan, China, and classification of cell type clusters for human nervous system based on the data from J. Craig Venter Institute. We show that RFEX offers simple yet powerful means of explaining RF classification at model, sample and feature levels, as well as providing guidance for testing and developing explainable and cost-effective operational prediction models.
Keywords
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Szabo, L., Kaiser Health News: Artificial intelligence is rushing into patient care—and could raise risks. Sci. Am. 24 December 2019
Kaufman, S., Rosset, S., Perlich, C.: Leakage in data mining: formulation, detection, and avoidance. ACM Trans. Knowl. Discov. Data 6(4), 1–21 (2012)
Dzindolet, M., Peterson, S., Pomranky, R., Pierce, L., Beck, H.: The role of trust in automation reliance. Int. J. Hum.-Comput. Stud. 58(6), 697–718 (2003)
Holm, E.: In defense of black box. Science 364(6435), 26–27 (2019)
Petkovic, D., Kobzik, L., Re, C.: Machine learning and deep analytics for biocomputing: call for better explainability. Pacific Symposium on Biocomputing Hawaii 23, 623–627 (2018)
Petkovic, D., Kobzik, L., Ganaghan, R.: AI ethics and values in biomedicine – technical challenges and solutions. In: Pacific Symposium on Biocomputing, Hawaii, 3–7 January (2020)
Vellido, A., Martin-Guerrero, J., Lisboa, P.: Making machine learning models interpretable. European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning; 25–27 April, Bruges, Belgium (2012)
Future of Life Institute: Asilomar AI Priciples. https://futureoflife.org/ai-principles/?cn-reloaded=1. Accessed 09 2020
Asociation of Computing machinery: Statement on Algorithmic Transparency and Accountability, 01 Dec 2017. https://www.acm.org/binaries/content/assets/public-policy/2017_usacm_statement_algorithms.pdf
OECD Principles on AI. https://www.oecd.org/going-digital/ai/principles/ Accessed 09 2020
Breiman, L.: Random forests. Mach. Learn. 45(1), 5–32 (2001)
Petkovic, D., Altman, R., Wong, M., Vigil, A.: Improving the explainability of Random Forest classifier - user centered approach. Pacific Symposium on Biocomputing. 23, 204–215 (2018)
L. Buturovic, M. Wong, G. Tang, R. Altman, D. Petkovic: “High precision prediction of functional sites in protein structures”, PLoS ONE 9(3): e91240. https://doi.org/10.1371/journal.pone.0091240
Okada, K., Flores, L., Wong, M., Petkovic, D.: Microenvironment-based protein function analysis by random forest. In: Proceedings of the ICPR (International Conference on Pattern Recognition), Stockholm (2014)
Yan, L., et al.: An Interpretable mortality prediction model for COVID-19 patients. Nature Mach. Intell. 2, pp. 283–288 (2020)
Aevermann, B., et al.: Cell type discovery using single cell transcriptomics: implications for ontological representation. Hum. Mol. Gene. 27(R1), R40–R47 (2018)
Aevermann, B., McCorrison, J., Venepally, P., et al.: Production of a preliminary quality control pipeline for single nuclei RNA-seq and its application in the analysis of cell type diversity of post-mortem human brain neocortex. In: Pacific Symposium on Biocomputing Proceedings, vol. 22, pp. 564–575, Hawaii, January 2017
Boldog, E., et al.: Transcriptomic and morphophysiological evidence for a specialized human cortical GABAergic cell type. Nat. Neurosci. 2018 21(9), 1185–1195. https://doi.org/10.1038/s41593-018-0205-2. Epub 2018 Aug 27
Yang, J., Petkovic, D.: Application of Improved Random Forest Explainability (Rfex 2.0) on Data from JCV Institute LaJolla, California, SFSU CS Department TR 19.01, 16 June 2019. https://cs.sfsu.edu/sites/default/files/technical-reports/RFEX%202%20JCVI_Jizhou%20Petkovic%20%2006-16-19_0.pdf
Alavi, A., Petkovic, D.: Improvements of Explainability of Random Forest Algorithms. SFSU CS Department TR TR 20.01, May 2020. https://cs.sfsu.edu/sites/default/files/technical-reports/Ali%20Alavi%20CER%20895%20RFEX%20May%202020.pdf
Olson, R.S., Cava, W., Mustahsan, Z., Varik, A., Moore, J.H.: Data-driven advice for applying machine learning to bioinformatics problems. Pac. Symp. Biocomput. 23, 192–203 (2018)
Liaw, A., Wiener, M.: Classification and regression by random forest. R News 2(3), 18–22 (2002). http://CRAN.R-project.org/doc/Rnews/
Solla, F., Tran, A., Bertoncelli, D., Musoff, C., Bertoncelli, C.M.: Why a P-value is not enough. Clin Spine Surg. 31(9), 385–388 (2018)
Barlaskar, S., Petkovic, D.: Applying Improved Random Forest Explainability (RFEX 2.0) on synthetic data. SFSU TR 18.01, 11/27/20181; with related toolkit at https://www.youtube.com/watch?v=neSVxbxxiCE
Acknowledgment
We are grateful to researchers from Huazhong University of Science and Technology, Wuhan, China for their prompt response to our inquiry for the COVID-19 data, and Dr. R. Scheuermann and B. Aevermann from JCVI for the data for our case study and their feedback. We are also grateful to Prof. Russ Altman, Stanford University, and Prof. Lester Kobzik (Harvard University) for their feedback and encouragement.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Petkovic, D., Alavi, A., Cai, D., Wong, M. (2021). Random Forest Model and Sample Explainer for Non-experts in Machine Learning – Two Case Studies. In: Del Bimbo, A., et al. Pattern Recognition. ICPR International Workshops and Challenges. ICPR 2021. Lecture Notes in Computer Science(), vol 12663. Springer, Cham. https://doi.org/10.1007/978-3-030-68796-0_5
Download citation
DOI: https://doi.org/10.1007/978-3-030-68796-0_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-68795-3
Online ISBN: 978-3-030-68796-0
eBook Packages: Computer ScienceComputer Science (R0)