Yazar "Ozdemir, O." seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim Yok
    Öğe
    Diagnosing The Breathing Sounds as COPD or Asthma
    (Institute of Electrical and Electronics Engineers Inc., 2022) Turkan, B.; Ates, A.G.; Ozdemir, O.; Sonmez, E.B.
    The aim of this research is to classify recorded chest sounds to distinguish among Asthma, Bronchiolitis, Bronchiectasis, COPD, Pneumonia and URTI diseases versus Healthy sound. That is, this paper introduces and challenges a seven- class problem using one of the few publicly available collection of sounds, the Respiratory Sound database from Kaggle. The performance of several deep learning algorithms has been compared and the Convolutional Neural Network architecture resulted in the most successful model. Unlike previous papers which worked on a subset of this database, this work proposes a more comprehensive seven-class challenge to distinguish among all diseases sampled in the database. The performance of several deep-learning algorithms has been compared and the best model is described in detail. © 2022 IEEE.
  • Küçük Resim Yok
    Öğe
    The effect of brand diversification on IPO returns: An examination of restaurant IPOs
    (Routledge, 2019) Ozdemir, O.; Erkmen, E.; Demirciftci, T.
    This study examines whether pre-IPO brand diversification is related to IPO returns in the restaurant industry. More precisely, the study examines whether brand-diversified restaurant firms experience a lower underpricing in their IPO relative to non-diversified (focused) restaurant firms. Second, the study investigates whether pre-IPO brand diversification affects long-run returns of restaurant IPOs. The sample of study is 106 restaurant firms that completed an IPO between 1981 and 2015. For primary analyses, t-test and ordinary least square regression are used. Findings of the study reveal that pre-IPO brand diversification is a significant firm attribute for a restaurant firm that goes through an IPO. Brand-diversified restaurant firms’ shares are more accurately priced by the investors, therefore they experience underpricing to a significantly lesser degree than focused restaurant firms. The study finds mixed evidence for the long-run returns. In most part, multivariate analyses suggest that pre-IPO brand diversification does not affect the long-run IPO returns of restaurant firms. © 2019, © 2019 Taylor & Francis.
  • Küçük Resim Yok
    Öğe
    Weighted Cross-Entropy for Unbalanced Data with Application on COVID X-ray images
    (Institute of Electrical and Electronics Engineers Inc., 2020) Ozdemir, O.; Sonmez, E.B.
    Since December 2019 the world is infected by COVID-19 or Coronavirus disease, which spreads very quickly, out of control. The high number of precautions for laboratory access, which need to be taken to contain the virus, together with the difficulties in running the gold standard test for COVID-19, result in a practical incapability to make early diagnosis. Recent advances in deep learning algorithms allow efficient implementation of computer-aided diagnosis. This paper investigates on the performance of a very well known residual network, ResNet50, and a lightweight Atrous CNN (ACNN) network using a Weighted Cross-entropy (WCE) loss function, to alleviate imbalance on COVID datasets. As a result, ResNet50 model initialized with pre-trained weights fine-tuned by ImageNet dataset and exploiting WCE achieved the state-of-the-art performance on COVIDXRay-5K test set, with a top balanced accuracy of 99.87%. © 2020 IEEE.