Multi-label ranking method based on positive class correlations
Source
Jordanian Journal of Computetrs and Information Technology
Issue
Vol. 6, Issue 4 (31 Dec. 2020), pp.377-391, 15 p.
Publisher
Princess Sumaya University for Technology
Publication Date
2020-12-31
Country of Publication
Jordan
No. of Pages
15
Main Subjects
Engineering & Technology Sciences (Multidisciplinary)
Abstract EN
Multi-label classification is a general type of classification that has attracted many researchers in the last two decades due to its applicability to many modern domains, such as scene classification, bioinformatics and text classification, among others.
This type of classification allows instances to be associated with more than one class label at the same time.
Class label ranking is a crucial problem in multi-label classification research, because it directly impacts the peiformance of the final classifiers, as labels with high ranks get a higher chance of being applied.
This paper presents a new multi-label ranking algorithm called Multi-label Ranking based on Positive Correlations among labels (MLR-PC).
MLR-PC captures positive correlations among labels to reduce the large search space and assigns the true rank per class label for multi-label classification problems.
More importantly, MLR-PC utilizes novel problem transformation methods that facilitate exploiting accurate positive correlations among labels.
This improves the predictive performance of the classification models derived.
Empirical results using different multi-label datasets and five evaluation metrics reveal that the MLR-PC is superior to other commonly existing classification algorithms.
American Psychological Association (APA)
al-Azaidah, Raid& Ahmad, Farzana Kabir& Muhsin, Muhammad Farhan Muhammad& al-Zubi, Wail Ahmad. 2020. Multi-label ranking method based on positive class correlations. Jordanian Journal of Computetrs and Information Technology،Vol. 6, no. 4, pp.377-391.
https://search.emarefa.net/detail/BIM-1415845
Modern Language Association (MLA)
al-Azaidah, Raid…[et al.]. Multi-label ranking method based on positive class correlations. Jordanian Journal of Computetrs and Information Technology Vol. 6, no. 4 (Dec. 2020), pp.377-391.
https://search.emarefa.net/detail/BIM-1415845
American Medical Association (AMA)
al-Azaidah, Raid& Ahmad, Farzana Kabir& Muhsin, Muhammad Farhan Muhammad& al-Zubi, Wail Ahmad. Multi-label ranking method based on positive class correlations. Jordanian Journal of Computetrs and Information Technology. 2020. Vol. 6, no. 4, pp.377-391.
https://search.emarefa.net/detail/BIM-1415845
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references : p. 389-391
Record ID
BIM-1415845