A tuned feed-forward deep neural network algorithm for effort estimation

ÖZTÜRK M. M.

JOURNAL OF EXPERIMENTAL & THEORETICAL ARTIFICIAL INTELLIGENCE, cilt.34, sa.2, ss.235-259, 2022 (SCI-Expanded, Scopus)

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 34 Sayı: 2
  • Basım Tarihi: 2022
  • Doi Numarası: 10.1080/0952813x.2021.1871664
  • Dergi Adı: JOURNAL OF EXPERIMENTAL & THEORETICAL ARTIFICIAL INTELLIGENCE
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Applied Science & Technology Source, Business Source Elite, Business Source Premier, Compendex, Computer & Applied Sciences, INSPEC, Psycinfo, zbMATH
  • Sayfa Sayıları: ss.235-259
  • Anahtar Kelimeler: Effort estimation, feed-forward deep neural network, hyperparameter tuning, machine learning
  • Süleyman Demirel Üniversitesi Adresli: Evet

Özet

Software effort estimation (SEE) is a software engineering problem that requires robust predictive models. To establish robust models, the most feasible configuration of hyperparameters of regression methods is searched. Although only a few works, which include hyperparameter optimisation (HO), have been done so far for SEE, there is not any comprehensive study including deep learning models. In this study, a feed-forward deep neural network algorithm (FFDNN) is proposed for software effort estimation. The algorithm relies on a binary-search-based method for finding hyperparameters. FFDNN outperforms five comparison algorithms in the experiment that uses two performance parameters. The results of the study suggest that: 1) Employing traditional methods such as grid and random search increases tuning time remarkably. Instead, sophisticated parameter search methods compatible with the structure of regression method should be developed; 2) The performance of SEE is enhanced when associated hyperparameter search method is devised according to the essentials of chosen deep learning approach; 3) Deep learning models achieve in competitive CPU time compared to the tree-based regression methods such as CART_DE8.