Dimensionality reduction in conic section function neural network


Creative Commons License

YILDIRIM T., ÖZYILMAZ L.

SADHANA-ACADEMY PROCEEDINGS IN ENGINEERING SCIENCES, cilt.27, ss.675-683, 2002 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 27
  • Basım Tarihi: 2002
  • Doi Numarası: 10.1007/bf02703358
  • Dergi Adı: SADHANA-ACADEMY PROCEEDINGS IN ENGINEERING SCIENCES
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.675-683
  • Anahtar Kelimeler: conic section function neural network, dimensionality reduction, hardware implementation, sensitivity analysis, UNIFIED FRAMEWORK
  • Yıldız Teknik Üniversitesi Adresli: Evet

Özet

This paper details how dimensionality can be reduced in conic section function neural networks (CSFNN). This is particularly important for hardware implementation of networks. One of the main problems to be solved when considering the hardware design is the high connectivity requirement. If the effect that each of the network inputs has on the network output after training a neural network is known, then some inputs can be removed from the network. Consequently, the dimensionality of the network, and hence, the connectivity and the training time can be reduced. Sensitivity analysis, which extracts the cause and effect relationship between the inputs and outputs of the network, has been proposed as a method to achieve this and is investigated for Iris plant, thyroid disease and ionosphere databases. Simulations demonstrate the validity of the method used.