A Brief Review on Deep Learning Based Software Vulnerability Detection

Alagoz Z. I., Akleylek S.

14th International Conference on Information Security and Cryptology, ISCTURKEY 2021, Ankara, Türkiye, 2 - 03 Aralık 2021, ss.143-148 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/iscturkey53027.2021.9654351
  • Basıldığı Şehir: Ankara
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.143-148
  • Anahtar Kelimeler: Convolutional Neural Network, Deep Learning, Deep Neural Network, Software Vulnerability
  • Ondokuz Mayıs Üniversitesi Adresli: Evet

Özet

Software vulnerabilities (SV) cause disastrous impact on information security in recent years. Higher cost and time consumption on manual detection methods lead to enormous number of increase in automatic SV detection techniques. Machine learning, deep learning (DL) and data mining methods are the most popular and efficient ones which also have advantage on analyzing performance results with use of available open-source softwares. This survey mainly focuses on the recent SV detection systems that use deep learning techniques. In this context, papers with significant impact on the literature are investigated, and deep learning methods, data sets and performance results are analyzed. Moreover, open problems and solution proposals are discussed.