Yazar "Temizceri, T." seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • Küçük Resim Yok
    Öğe
    An Online Shoppers Purchasing Intention Model Based on Ensemble Learning
    (Institute of Electrical and Electronics Engineers Inc., 2023) Karakaya, A.; Karakaya, I.; Temizceri, T.
    The sales trends and the visitors' purchase intention of e-commerce are crucial issues for companies. Therefore, companies want to know whether products are purchased based on customer visits. This study proposes an ensemble learning-based model to analyze customers' purchase intentions for e-commerce companies. It is possible to create a more accurate model using several machine learning methods within a system, which is the ensemble learning approach. The prediction performance of the proposed model is evaluated by different metrics, such as accuracy, precision, recall, AUC, and F-score. This paper aims to improve the purchasing rates of online shopping sites by determining the rate of visitors who intend to purchase but leave the site without taking action by using information such as the click flow of visitors. © 2023 IEEE.
  • Küçük Resim Yok
    Öğe
    A Variable Block Insertion Heuristic for the Energy-Efficient Permutation Flowshop Scheduling with Makespan Criterion
    (Springer Science and Business Media Deutschland GmbH, 2021) Tasgetiren, M.F.; Oztop, H.; Pan, Q.-K.; Ornek, M.A.; Temizceri, T.
    Permutation flow shop scheduling problem is a well-known problem in the scheduling literature. Even though various multi-objective permutation flowshop scheduling problems have been studied in the literature, energy consumption consideration in scheduling is still very seldom. In this paper, we consider a bi-objective permutation flowshop scheduling problem with the objectives of minimizing the total energy consumption and the makespan. We present a bi-objective mixed integer programming model for the problem applying a speed-scaling approach. Then, we employ the augmented ? -constraint method to generate the Pareto-optimal solution sets for small-sized instances. For larger instances, we use the augmented ? -constraint method with a time limit on CPLEX solver to approximate the Pareto frontiers. We also propose a heuristic approach, which employs a very recent variable block insertion heuristic algorithm. In order to evaluate performance of the proposed algorithm, we have carried out detailed computational experiments using well-known benchmarks from the literature. First, we present the performance of the proposed algorithm on small-sized problems; then, we show that the proposed algorithm is very effective to solve larger problems as compared with the time-limited CPLEX. © 2020, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG.