We present a sensitivity analysis-based method for explaining prediction models that can be applied to any type of classification or regression model. Its advantage over existing general methods is that all subsets of input features are perturbed, so interactions and redundancies between features are taken into account. Furthermore, when explaining an additive model, the method is equivalent to commonly used additive model-specific methods. We illustrate the method’s usefulness with examples from artificial and real-world data sets and an empirical analysis of running times. Results from a controlled experiment with 122 participants suggest that the method’s explanations improved the participants’ understanding of the model.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
Tax calculation will be finalised during checkout.
Achen CH (1982) Intepreting and Using Regression. Sage Publications, Thousand Oaks
Allahyari H, Lavesson N (2011) User-oriented assessment of classification model understandability. In: Proceedings of the 11th Scandinavian conference on artificial intelligence, SCAI 2011, pp 11–19
Becker B, Kohavi R, Sommerfield D (1997) Visualizing the simple Bayesian classier. KDD workshop on issues in the integration of data mining and data visualization
Bhattacharya S, Xu D, Kumar K (2011) An ANN-based auditor decision support system using Benford’s law. Decis Support Syst 50(3):576–584
Bhattacharyya S, Jha S, Tharakunnel K, Westland JC (2011) Data mining for credit card fraud: a comparative study. Decis Support Syst 50(3):602–613
Blanchard J, Guillet F, Briand H (2007) Interactive visual exploration of association rules with rule-focusing methodology. Knowl Inf Syst 13:43–75
Castro J, Gómez D, Tejada J (2009) Polynomial calculation of the shapley value based on sampling. Comput Oper Res 36(5):1726–1730
De Falco I, Della Cioppa A (2005) An evolutionary approach for automatically extracting intelligible classification rules. Knowl Inf Syst 7:179–201
Frank A, Asuncion A (2011) Uci machine learning repository
Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P, Witten IH (2009) The weka data mining software: an update. SIGKDD Explor Newsl 11(1):10–18
Huang Z, Chen H, Hsu CJ, Chen WH, Wu S (2004) Credit rating analysis with support vector machines and neural networks: a market comparative study. Decis Support Syst 37(4):543–558
Huysmans J, Dejaeger K, Mues C, Vanthienen J, Baesens B (2011) An empirical evaluation of the comprehensibility of decision table, tree and rule based predictive models. Decis Support Syst 51(1):141–154
Jaeckel P (2002) Monte Carlo methods in finance. Wiley, New York
Jakulin A, Možina M, Demšar J, Bratko I, Zupan B (2005) Nomograms for visualizing support vector machines. KDD ’05: 11th ACM SIGKDD, ACM, pp 108–117
Kattan MW, Eastham JA, Stapleton AM, Wheeler TM, Scardino PT (1998) A preoperative nomogram for disease recurrence following radical prostatectomy for prostate cancer. J Natl Cancer Inst 90:766–771
Knuth DE (1998) The art of computer programming, volume 2: seminumerical algorithms. Addison-Wesley, Boston
Kononenko I (1993) Inductive and bayesian learning in medical diagnosis. Appl Artif Intell 7:317–337
Lee S (2010) Using data envelopment analysis and decision trees for efficiency analysis and recommendation of B2C controls. Decis Support Syst 49(4):486–497
Lemaire V, Feraud R, Voisine N (2008) Contact personalization using a score understanding method. In: International joint conference on neural networks (IJCNN)
Lim BY, Dey AK, Avrahami D (2009) Why and why not explanations improve the intelligibility of context-aware intelligent systems. In: Proceedings of the 27th international conference on Human factors in computing systems, CHI ’09, ACM, New York, NY, USA, pp 2119–2128
Lubsen J, Pool J, van der Does E (1978) A practical device for the application of a diagnostic or prognostic function. Methods Inf Med 17:127–129
Melli G (n.d.) The datgen dataset generator. http://www.datasetgenerator.com
Možina M, Demšar J, Kattan M, Zupan B (2004) Nomograms for visualization of naive Bayesian classifier. PKDD 2004, Springer, pp 337–348
Robnik-Šikonja M, Kononenko I (2008) Explaining classifications for individual instances. IEEE TKDE 20:589–600
Shapley LS (1953) A value for n-person games, vol II of Contributions to the theory of games. Princeton University Press, Princeton
Szafron D, Poulin B, Eisner R, Lu P, Greiner R, Wishart D, Fyshe A, Pearcy B, Macdonell C, Anvik J (2006) Visual explanation of evidence in additive classifiers. In: Proceedings of innovative applications of artificial intelligence
Štrumbelj E, Bosnić Z, Zakotnik B, Grašič-Kuhar C, Kononenko I (2010) Explanation and reliability of breast cancer recurrence predictions. Knowl Inf Syst 24(2):305–324
Štrumbelj E, Kononenko I (2010) An efficient explanation of individual classifications using game theory. J Mach Learn Res 11:1–18
Štrumbelj E, Kononenko I (2011) A general method for visualizing and explaining black-box regression models. In: Dobnikar A, Lotric U, Ster B (eds) ICANNGA (2), vol 6594 of Lecture notes in computer science. Springer, Berlin, pp 21–30
Welford BP (1962) Note on a method for calculating corrected sums of squares and products. Technometrics 4(3):419–420
About this article
Cite this article
Štrumbelj, E., Kononenko, I. Explaining prediction models and individual predictions with feature contributions. Knowl Inf Syst 41, 647–665 (2014). https://doi.org/10.1007/s10115-013-0679-x
- Knowledge discovery
- Data mining
- Decision support