Efficient Sequential Covering Strategy for Classification Rules Mining using a Discrete Equilibrium Optimization Algorithm

Simple item page
dc.contributor.authorMALIK, Mohamed MAhdi
dc.contributor.authorHAOUASSI, Hichem
dc.date.accessioned2024-02-13T09:01:01Z
dc.date.available2024-02-13T09:01:01Z
dc.date.issued2022-10
dc.descriptionhttps://www.sciencedirect.com/science/article/pii/S1319157821002500
dc.description.abstractRule-based classification is one of the important tasks in data mining due to its wide applications, particularly in the domains that need to interpret the classification decision such as medical diagnosis. The rule-based classification is a combination of the classification and association rule mining fields which aims at building interpretable classifiers by means of classification rules. This paper presents a novel and efficient sequential covering strategy for Classification Rule Mining to improve the interpretability of classifiers using a Discrete Equilibrium Optimization Algorithm called DEOA-CRM. Our approach benefits from the advantages of associative classification and population-based intelligence. It is inspired by the recent meta-heuristic equilibrium optimization algorithm. New discrete operators defined enable our approach to avoid local solutions and find global ones, improving the exploration and exploitation power in the search space. The proposed DEOA-CRM is tested on a total of 12 test data sets of various sizes and benchmarked with four recent and well-known rule-based classification mining algorithms. The obtained results confirm the efficiency of our algorithm in three chosen measures. Our approach fully deserves its use for classification rules generation to help decision-makers generate accurate and interpretable models. Keywords: Rule-based classification; Classification rules; Population-based optimization; Discrete equilibrium optimization algorithm; Classifier’s interpretability
dc.identifier.citationMohamed Mahdi MALIK, Hichem HAOUASSI, Efficient sequential covering strategy for classification rules mining using a discrete equilibrium optimization algorithm, Journal of King Saud University - Computer and Information Sciences, Volume 34, Issue 9, 2022, Pages 7559-7569, ISSN 1319-1578, https://doi.org/10.1016/j.jksuci.2021.08.032.
dc.identifier.urihttp://dspace.univ-khenchela.dz:4000/handle/123456789/605
dc.language.isoen
dc.publisherJournal of King Saud University - Computer and Information Sciences
dc.relation.ispartofseriesVolume 34; Issue 9
dc.titleEfficient Sequential Covering Strategy for Classification Rules Mining using a Discrete Equilibrium Optimization Algorithm
dc.typeArticle
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
Efficient Sequential Covering Strategy for CRM using a DEOA_final_2021_FIRST-PAGE.pdf
Size:
46.83 KB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed to upon submission
Description:
Download
Collections
scientific publication