Novel class detection using hybrid ensemble

Diptangshu Pandit, Li Zhang, Kamlesh Mistry, Richard Jiang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this research, we propose a hybrid meta-classifier for novel class detection. It is able to efficiently detect the arrival of novel unseen classes as well as tackle real-time data stream classification. Specifically, the proposed hybrid meta-classifler includes three ensemble models, i.e. class-specific, cluster-specific and complementary boosting ensemble classifiers. Distinctive training strategies are also proposed for the generation of effective and diversified ensemble classifiers. The weights of the above ensemble models and the threshold of the novel class confidence are subsequently optimized using a modified Firefly Algorithm, to enhance performance. The above proposed ensemble and optimization algorithms cooperate with each other to conduct the detection of novel unseen classes. Several UCI databases are employed for evaluation, i.e. The KDD Cup, Image Segmentation, Soybean Large, Glass and Iris databases. In comparison with the baseline meta-algorithms such as Boosting, Bagging and Stacking, our approach shows significantly enhanced performance with the increment of the number of novel classes for all the test data sets, which poses great challenges to the existing baseline ensemble methods.

Original languageEnglish
Title of host publicationProceedings of 2020 International Conference on Machine Learning and Cybernetics, ICMLC 2020
Place of PublicationPiscataway, NJ
PublisherIEEE
Pages267-272
Number of pages6
ISBN (Electronic)9780738124261, 9781665419437
ISBN (Print)9781665430074
DOIs
Publication statusPublished - 2 Dec 2020
Event19th International Conference on Machine Learning and Cybernetics, ICMLC 2020 - Virtual, Online
Duration: 4 Dec 2020 → …

Publication series

NameProceedings - International Conference on Machine Learning and Cybernetics
Volume2020-December
ISSN (Print)2160-133X
ISSN (Electronic)2160-1348

Conference

Conference19th International Conference on Machine Learning and Cybernetics, ICMLC 2020
CityVirtual, Online
Period4/12/20 → …

Fingerprint

Dive into the research topics of 'Novel class detection using hybrid ensemble'. Together they form a unique fingerprint.

Cite this