Engineering

Permanent URI for this communityhttps://hdl.handle.net/11693/115572

The Bilkent University Open Access Articles collection consists of scholarly articles written by Bilkent-affiliated authors that are made available through BUIR under the Bilkent Faculty Open Access Policy, or under related publisher agreements.

Version details are supplied for each paper in the collection:

  • Original manuscript: author's manuscript prior to formal peer review
  • Author's final manuscript: final author's manuscript post peer review, without publisher's formatting or copy editing
  • Final published version: final published article, as it appeared in a journal, conference proceedings, or other formally published context (this version appears here only if allowable under publisher's policy)

Post peer-reviewed scholarly articles are available through BUIR collections, such as those for departments, labs, and research centers.

Browse

Browsing Engineering by Author "Abadifard, Soheil"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Thumbnail Image
    ItemOpen Access
    DynED: dynamic ensemble diversification in data stream classification
    (Association for Computing Machinery, 2023-10-23) Abadifard, Soheil; Gheibuni, Sanaz; Bakhshi, Sepehr; Can, Fazlı
    Ensemble methods are commonly used in classification due to their remarkable performance. Achieving high accuracy in a data stream environment is a challenging task considering disruptive changes in the data distribution, also known as concept drift. A greater diversity of ensemble components is known to enhance predic tion accuracy in such settings. Despite the diversity of components within an ensemble, not all contribute as expected to its overall performance. This necessitates a method for selecting components that exhibit high performance and diversity. We present a novel ensemble construction and maintenance approach based on MMR (Maximal Marginal Relevance) that dynamically combines the diver sity and prediction accuracy of components during the process of structuring an ensemble. The experimental results on both four real and 11 synthetic datasets demonstrate that the proposed approach (DynED) provides a higher average mean accuracy compared to the five state-of-the-art baselines.