Abstract
In some sensitive domains where data imperfections are present, standard classification techniques reach their limits. To avoid misclassification that has serious consequences, recent works propose cautious classification algorithms to handle the problem. Despite of the presence of uncertainty, a point prediction classifier is forced to decide which single class to associate to a sample. In such a case, a cautious classifier proposes the appropriate subset of candidate classes that can be assigned to the sample in the presence of imperfect information. On the other hand, cautiousness should not override relevance and a trade-off has to be made between these two criteria. Among the existing cautious classifiers, two classifiers propose to manage this trade-off in the decision step of the classifier algorithm by the mean of a parametrized objective function. The first one is the non-deterministic classifier (ndc) proposed within the framework of probability theory and the second one is eclair (evidential classifier based on imprecise relabelling) proposed within the framework of belief functions. The theoretical aim of the mentioned parameter is to control the size of predictions for both classifiers. This paper proposes to study this parameter in order to select the “best" value in a classification task. First the gain for each prediction candidate is studied related to the values of the hyper-parameter. In the illustration section, we propose a method to choose this hyper-parameter base on the training data and we show the classification results on randomly generated data and we present some comparisons with two other imprecise classifiers on 11 UCI datasets based on five measures of imprecise classification performances used in the state of the art.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Abellan, J., Masegosa, A.R.: Imprecise classification with credal decision trees. Int. J. Uncertainty Fuzziness Knowl. Based Syst. 20(05), 763–787 (2012)
Couso, I., Sánchez, L.: Machine learning models, epistemic set-valued data and generalized loss functions: an encompassing approach. Inf. Sci. 358, 129–150 (2016)
Coz, J.J.d., Díez, J., Bahamonde, A.: Learning nondeterministic classifiers. J. Mach. Learn. Res. 10, 2273–2293 (2009)
Imoussaten, A., Jacquin, L.: Cautious classification based on belief functions theory and imprecise relabelling. Int. J. Approximate Reasoning 142, 130–146 (2022)
Jacquin, L., Imoussaten, A., Trousset, F., Montmain, J., Perrin, D.: Evidential classification of incomplete data via imprecise relabelling: application to plastic sorting. In: Ben Amor, N., Quost, B., Theobald, M. (eds.) Scalable Uncertainty Management, pp. 122–135. Springer International Publishing, Cham (2019). https://doi.org/10.1007/978-3-030-35514-2_10
Jacquin, L., Imoussaten, A., Trousset, F., Perrin, D., Montmain, J.: Control of waste fragment sorting process based on MIR imaging coupled with cautious classification. Resour. Conserv. Recycl. 168, 105258 (2021)
Ma, L., Denoeux, T.: Partial classification in the belief function framework. Knowl.-Based Syst. 214, 106742 (2021)
Quost, B., Masson, M.-H., Destercke, S.: Dealing with Atypical Instances in Evidential Decision-Making. In: Davis, J., Tabia, K. (eds.) SUM 2020. LNCS (LNAI), vol. 12322, pp. 217–225. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58449-8_15
Sanchez, L., Couso, I.: A framework for learning fuzzy rule-based models with epistemic set-valued data and generalized loss functions. Int. J. Approximate Reasoning 92, 321–339 (2018)
Tsoumakas, G., Vlahavas, I.: Random k-Labelsets: an ensemble method for multilabel classification. In: Kok, J.N., Koronacki, J., Mantaras, R.L., Matwin, S., Mladenič, D., Skowron, A. (eds.) ECML 2007. LNCS (LNAI), vol. 4701, pp. 406–417. Springer, Heidelberg (2007). https://doi.org/10.1007/978-3-540-74958-5_38
Yang, G., Destercke, S., Masson, M.H.: The costs of indeterminacy: how to determine them? IEEE Trans. Cybern. 47(12), 4316–4327 (2016)
Zaffalon, M.: A credal approach to Naive classification. In: ISIPTA. vol. 99, pp. 405–414 (1999)
Zaffalon, M.: Statistical inference of the Naive credal classifier. In: ISIPTA. vol. 1, pp. 384–393 (2001)
Zaffalon, M., Corani, G., Mauá, D.: Evaluating credal classifiers by utility-discounted predictive accuracy. Int. J. Approximate Reasoning 53(8), 1282–1301 (2012)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Imoussaten, A. (2022). Choosing the Decision Hyper-parameter for Some Cautious Classifiers. In: Ciucci, D., et al. Information Processing and Management of Uncertainty in Knowledge-Based Systems. IPMU 2022. Communications in Computer and Information Science, vol 1602. Springer, Cham. https://doi.org/10.1007/978-3-031-08974-9_61
Download citation
DOI: https://doi.org/10.1007/978-3-031-08974-9_61
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-08973-2
Online ISBN: 978-3-031-08974-9
eBook Packages: Computer ScienceComputer Science (R0)