Arşiv logosu
  • Türkçe
  • English
  • Giriş
    Yeni kullanıcı mısınız? Kayıt için tıklayın. Şifrenizi mi unuttunuz?
Arşiv logosu
  • Koleksiyonlar
  • Sistem İçeriği
  • Analiz
  • Talep/Soru
  • Türkçe
  • English
  • Giriş
    Yeni kullanıcı mısınız? Kayıt için tıklayın. Şifrenizi mi unuttunuz?
  1. Ana Sayfa
  2. Yazara Göre Listele

Yazar "Parlak, Bekir" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
Sayfa Başına Sonuç
Sıralama seçenekleri
  • Yükleniyor...
    Küçük Resim
    Öğe
    A novel filter feature selection method for text classification: Extensive Feature Selector
    (SAGE Publications, Early Access) Parlak, Bekir; Uysal, Alper Kürşat
    As the huge dimensionality of textual data restrains the classification accuracy, it is essential to apply feature selection (FS) methods as dimension reduction step in text classification (TC) domain. Most of the FS methods for TC contain several number of probabilities. In this study, we proposed a new FS method named as Extensive Feature Selector (EFS), which benefits from corpus-based and classbased probabilities in its calculations. The performance of EFS is compared with nine well-known FS methods, namely, Chi-Squared (CHI2), Class Discriminating Measure (CDM), Discriminative Power Measure (DPM), Odds Ratio (OR), Distinguishing Feature Selector (DFS), Comprehensively Measure Feature Selection (CMFS), Discriminative Feature Selection (DFSS), Normalised Difference Measure (NDM) and Max–Min Ratio (MMR) using Multinomial Naive Bayes (MNB), Support-Vector Machines (SVMs) and k-Nearest Neighbour (KNN) classifiers on four benchmark data sets. These data sets are Reuters-21578, 20-Newsgroup, Mini 20-Newsgroup and Polarity. The experiments were carried out for six different feature sizes which are 10, 30, 50, 100, 300 and 500. Experimental results show that the performance of EFS method is more successful than the other nine methods in most cases according to microF1 and macro-F1 scores.
  • [ X ]
    Öğe
    Hybrid structures: innovative governance, judicial and sociological approaches
    (Springer, 2015) Aliu, Armando; Parlak, Bekir; Aliu, Dorian
    The aim of this study is to examine the new trends in the hybridity research area and clarify the convergence of interests among state actors, private actors and civil society actors. Hybridity is perceived as a multidimensional phenomenon and new paradigm in various industries. The effective collaboration among public sector-private sector-civil society can be attained likelihood with taking into account regional governance and multilevel governance. Hybridity at global governance level covers decentration (supra: centralization and infra: decentralization; or politically/socioeconomically: quasi-decentralization and culturally: denationalization) which includes the nexus of voice (democratic participation) and entitlement (legal/social rights and duties). In this framework, this study explores state and non-state interactions at multiple levels and attempts to clarify how hybridization provides triple win solution for state actors, private actors and civil society actors related issues in realm of theory/praxis dichotomy. Through enhancing legitimacy and effectiveness of the activities and efforts of non-state actors in the framework of (quasi)indirect centralization process, states ensure reciprocal understanding. In this study, constructivism was followed as paradigmatic research method.

| Çanakkale Onsekiz Mart Üniversitesi | Kütüphane | Açık Erişim Politikası | Rehber | OAI-PMH |

Bu site Creative Commons Alıntı-Gayri Ticari-Türetilemez 4.0 Uluslararası Lisansı ile korunmaktadır.


Çanakkale Onsekiz Mart Üniversitesi, Çanakkale, TÜRKİYE
İçerikte herhangi bir hata görürseniz lütfen bize bildirin

DSpace 7.6.1, Powered by İdeal DSpace

DSpace yazılımı telif hakkı © 2002-2025 LYRASIS

  • Çerez Ayarları
  • Gizlilik Politikası
  • Son Kullanıcı Sözleşmesi
  • Geri Bildirim