Home > Research > Publications & Outputs > Transparent Deep Rule-Based Classifiers

Links

Text available via DOI:

View graph of relations

Transparent Deep Rule-Based Classifiers

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNChapter (peer-reviewed)peer-review

Published

Standard

Transparent Deep Rule-Based Classifiers. / Angelov, P.P.; Gu, X.

Empirical Approach to Machine Learning. ed. / Plamen Angelov; Xiaowei Gu. Vol. 800 Springer-Verlag, 2019. p. 223-245 (Studies in Computational Intelligence; Vol. 800).

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNChapter (peer-reviewed)peer-review

Harvard

Angelov, PP & Gu, X 2019, Transparent Deep Rule-Based Classifiers. in P Angelov & X Gu (eds), Empirical Approach to Machine Learning. vol. 800, Studies in Computational Intelligence, vol. 800, Springer-Verlag, pp. 223-245. https://doi.org/10.1007/978-3-030-02384-3_9

APA

Angelov, P. P., & Gu, X. (2019). Transparent Deep Rule-Based Classifiers. In P. Angelov, & X. Gu (Eds.), Empirical Approach to Machine Learning (Vol. 800, pp. 223-245). (Studies in Computational Intelligence; Vol. 800). Springer-Verlag. https://doi.org/10.1007/978-3-030-02384-3_9

Vancouver

Angelov PP, Gu X. Transparent Deep Rule-Based Classifiers. In Angelov P, Gu X, editors, Empirical Approach to Machine Learning. Vol. 800. Springer-Verlag. 2019. p. 223-245. (Studies in Computational Intelligence). https://doi.org/10.1007/978-3-030-02384-3_9

Author

Angelov, P.P. ; Gu, X. / Transparent Deep Rule-Based Classifiers. Empirical Approach to Machine Learning. editor / Plamen Angelov ; Xiaowei Gu. Vol. 800 Springer-Verlag, 2019. pp. 223-245 (Studies in Computational Intelligence).

Bibtex

@inbook{f5cbf13ac7b64b6d84549c2e9a3a1785,
title = "Transparent Deep Rule-Based Classifiers",
abstract = "In this chapter, a new type of deep rule-based (DRB) classifier with a multi-layer architecture is presented for image classification, which combines the computer vision techniques with a massively parallel set of zero-order fuzzy rules as its learning engine. With its prototype-based nature, the DRB classifiers are able to identify a transparent and human-understandable fuzzy rule-based (FRB) system structure from the data through an autonomous, non-iterative, non-parametric and highly parallel online learning process, and offer extremely high classification accuracy. The DRB classifier can start “from scratch”, and conduct classification from the very first image of each class in the same way as humans do. The DRB classifier can also learn in a semi-supervised mode initialized with only a small proportion of the labelled data and continue in a fully unsupervised mode after that. The ability of semi-supervised learning further allows the DRB classifier to learn new classes actively without human experts{\textquoteright} involvement. Thanks to the prototype-based nature of the DRB classifier, it is free from prior assumptions about the type of the data distribution, their random or deterministic nature, and there are no requirements to make ad hoc decisions. Its supervised and semi-supervised learning processes are fully transparent and human-interpretable. The semi-supervised DRB classifiers can perform classification on out-of-sample images and also support recursive online training on a sample-by-sample basis or a batch-by-batch basis. {\textcopyright} 2019, Springer Nature Switzerland AG.",
author = "P.P. Angelov and X. Gu",
year = "2019",
doi = "10.1007/978-3-030-02384-3_9",
language = "English",
isbn = "9783030023836",
volume = "800",
series = "Studies in Computational Intelligence",
publisher = "Springer-Verlag",
pages = "223--245",
editor = "Plamen Angelov and Xiaowei Gu",
booktitle = "Empirical Approach to Machine Learning",

}

RIS

TY - CHAP

T1 - Transparent Deep Rule-Based Classifiers

AU - Angelov, P.P.

AU - Gu, X.

PY - 2019

Y1 - 2019

N2 - In this chapter, a new type of deep rule-based (DRB) classifier with a multi-layer architecture is presented for image classification, which combines the computer vision techniques with a massively parallel set of zero-order fuzzy rules as its learning engine. With its prototype-based nature, the DRB classifiers are able to identify a transparent and human-understandable fuzzy rule-based (FRB) system structure from the data through an autonomous, non-iterative, non-parametric and highly parallel online learning process, and offer extremely high classification accuracy. The DRB classifier can start “from scratch”, and conduct classification from the very first image of each class in the same way as humans do. The DRB classifier can also learn in a semi-supervised mode initialized with only a small proportion of the labelled data and continue in a fully unsupervised mode after that. The ability of semi-supervised learning further allows the DRB classifier to learn new classes actively without human experts’ involvement. Thanks to the prototype-based nature of the DRB classifier, it is free from prior assumptions about the type of the data distribution, their random or deterministic nature, and there are no requirements to make ad hoc decisions. Its supervised and semi-supervised learning processes are fully transparent and human-interpretable. The semi-supervised DRB classifiers can perform classification on out-of-sample images and also support recursive online training on a sample-by-sample basis or a batch-by-batch basis. © 2019, Springer Nature Switzerland AG.

AB - In this chapter, a new type of deep rule-based (DRB) classifier with a multi-layer architecture is presented for image classification, which combines the computer vision techniques with a massively parallel set of zero-order fuzzy rules as its learning engine. With its prototype-based nature, the DRB classifiers are able to identify a transparent and human-understandable fuzzy rule-based (FRB) system structure from the data through an autonomous, non-iterative, non-parametric and highly parallel online learning process, and offer extremely high classification accuracy. The DRB classifier can start “from scratch”, and conduct classification from the very first image of each class in the same way as humans do. The DRB classifier can also learn in a semi-supervised mode initialized with only a small proportion of the labelled data and continue in a fully unsupervised mode after that. The ability of semi-supervised learning further allows the DRB classifier to learn new classes actively without human experts’ involvement. Thanks to the prototype-based nature of the DRB classifier, it is free from prior assumptions about the type of the data distribution, their random or deterministic nature, and there are no requirements to make ad hoc decisions. Its supervised and semi-supervised learning processes are fully transparent and human-interpretable. The semi-supervised DRB classifiers can perform classification on out-of-sample images and also support recursive online training on a sample-by-sample basis or a batch-by-batch basis. © 2019, Springer Nature Switzerland AG.

U2 - 10.1007/978-3-030-02384-3_9

DO - 10.1007/978-3-030-02384-3_9

M3 - Chapter (peer-reviewed)

SN - 9783030023836

VL - 800

T3 - Studies in Computational Intelligence

SP - 223

EP - 245

BT - Empirical Approach to Machine Learning

A2 - Angelov, Plamen

A2 - Gu, Xiaowei

PB - Springer-Verlag

ER -