Fakülteler / Faculties

Permanent URI for this communityhttps://hdl.handle.net/11727/1395

Browse

Filters

2018 - 201912020 - 20211

Settings

search.filters.applied.f.rights: search.filters.rights.info:eu-repo/semantics/openAccessSubject: search.filters.subject.Classification

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Investigation of some machine learning algorithms in fish age classification
    (2021) Benzer, Semra; Garabaghi, Farid Hassanbaki; Benzer, Recep; Mehr, Homay Danaei; 0000-0002-5339-0554; A-5050-2014
    Marine and freshwater scientists use fish scales, vertebrae, otoliths and length-weights values to estimate fish age because reliable fish age estimation plays a very important role in fish stock management. The advances in technology and the widespread use of artificial intelligence have revealed the use of traditional observations and techniques in the fishing industry. The aim of this study was to evaluate the effectiveness of three disesteemed machine learning algorithms (NB, J48 DT, RF) in comparison with ANNs which has been widely used in such studies in the literature. In culmination, all three algorithms outperformed ANNs and can be considered as alternatives in case of coming across noisy and non-linear datasets. Moreover, among these three algorithms J48 DT and RF showed exceptional performance where the data for specific fish age groups weren't abundant.
  • Thumbnail Image
    Item
    Sparsity-driven weighted ensemble classifier
    (2018) Erdem, Hamit; Ozgur, Atilla; Nar, Fatih
    In this study, a novel sparsity-driven weighted ensemble classifier (SDWEC) that improves classification accuracy and minimizes the number of classifiers is proposed. Using pre-trained classifiers, an ensemble in which base classifiers votes according to assigned weights is formed. These assigned weights directly affect classifier accuracy. In the proposed method, ensemble weights finding problem is modeled as a cost function with the following terms: (a) a data fidelity term aiming to decrease misclassification rate, (b) a sparsity term aiming to decrease the number of classifiers, and (c) a non-negativity constraint on the weights of the classifiers. As the proposed cost function is non-convex thus hard to solve, convex relaxation techniques and novel approximations are employed to obtain a numerically efficient solution. Sparsity term of cost function allows trade-off between accuracy and testing time when needed. The efficiency of SDWEC was tested on 11 datasets and compared with the state-of-the art classifier ensemble methods. The results show that SDWEC provides better or similar accuracy levels using fewer classifiers and reduces testing time for ensemble.