3D Teaching-Learning-Based-Optimization Algorithm and Rating Problem

AR Y., KARTLI N., YİĞİT SERT S.

6th International Conference on Problems of Cybernetics and Informatics, PCI 2025, Baku, Azerbaycan, 25 - 28 Ağustos 2025, (Tam Metin Bildiri) identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/pci66488.2025.11219745
  • Basıldığı Şehir: Baku
  • Basıldığı Ülke: Azerbaycan
  • Anahtar Kelimeler: 3D problems, metaheuristic algorithms, optimization, teaching-learning-based optimization
  • Ankara Üniversitesi Adresli: Evet

Özet

It is known that metaheuristic algorithms are effective in finding approximate solutions to many optimization problems. However, these algorithms typically require adaptation before they can be directly applied to the specific problem. One of the successful metaheuristic optimization algorithms is the Teaching-Learning-Based Optimization (TLBO) algorithm. On the other hand, rating prediction refers to estimating the score a user would assign to an item (e.g., a movie, product, or book) based on their past interactions, and it is a widely studied problem in the field of recommender systems. In this study, we address the optimization problem of a matrix-dependent objective function. We demonstrate that the classical TLBO algorithm can be adapted for the problem and propose a 3D version of this algorithm. We then demonstrate that the proposed 3D TLBO algorithm can be applied to the rating prediction problem. Experimental results on simple examples indicate that this algorithm can produce effective outcomes.