Browsing by Author "Ozmen, Mihrimah"

Now showing 1 - 3 of 3

Citation - WoS: 1
Citation - Scopus: 2
Accelerated Artificial Bee Colony Optimization for Cost-Sensitive Neural Networks in Multi-Class Problems
(Wiley, 2025) Hacilar, Hilal; Dedeturk, Bilge Kagan; Ozmen, Mihrimah; Celik, Mehlika Eraslan; Gungor, Vehbi Cagri; 01. Abdullah Gül University; 02. 04. Bilgisayar Mühendisliği; 02. Mühendislik Fakültesi
Metaheuristics are advanced problem-solving techniques that develop efficient algorithms to address complex challenges, while neural networks are algorithms inspired by the structure and function of the human brain. Combining these approaches enables the resolution of complex optimization problems that traditional methods struggle to solve. This study presents a novel approach integrating the ABC algorithm with ANNs for weight optimization. The method is further enhanced by vectorization and parallelization techniques on both CPU and GPU to improve computational efficiency. Additionally, this study introduces a cost-sensitive fitness function tailored for multi-class classification to optimize results by considering relationships between target class levels. It validates these advancements in two critical applications: network intrusion detection and earthquake damage estimation. Notably, this study makes a significant contribution to earthquake damage assessment by leveraging machine learning algorithms and metaheuristics to enhance predictive models and decision-making in disaster response. By addressing the dynamic nature of earthquake damage, this research fills a critical gap in existing models and broadens the understanding of how machine learning and metaheuristics can improve disaster response strategies. In both domains, the ABC-ANN implementation yields promising results, particularly in earthquake damage estimation, where the cost-sensitive approach demonstrates satisfactory outcomes in macro-F1 and accuracy. The best results for macro-F1, weighted-F1, and overall accuracy provides best results with the UNSW-NB15 and earthquake datasets, showing values of 64%, 72%, 68%, and 60%, 80%, and 79%, respectively. Comparative performance evaluations reveal that the proposed parallel ABC-ANN model, incorporating the novel cost-sensitive fitness function and enhanced by vectorization and parallelization techniques, significantly reduces training time and outperforms state-of-the-art methods in terms of macro-F1 and accuracy in both network intrusion detection and earthquake damage estimation.
Determining the Priority Waste in Aluminum Manufacturing Sector Using the SMSA-2 Method: A Case Study of Kayseri
(Computers and Industrial Engineering dessouky@usc.edu, 2014) Kızılkaya Aydoğan, Emel Kizilkaya; Ates, Nuray; Uzal, Niǧmet; Ozmen, Mihrimah; 01. Abdullah Gül University; 02.03. İnşaat Mühendisliği; 02. Mühendislik Fakültesi
Small and medium-sized enterprises (SMEs) constitute a major part of the Turkish economy, accounting for a large proportion of the country's businesses and total employment. Although the SMEs are known as important contributors to environmental pollution, the relative contribution of SMEs to the total environmental impacts of industrial is unknown. The most important environmental issues related with aluminum industries are emission of gases, wastewater and solid wastes from aluminum production. In multi-criteria decision making (MCDM) problems in some situations, decision makers (DMs) don't or can't express their preferences obviously. In these situations for decision making, stochastic multi-criteria acceptability analysis (SMAA-2) can be applied. In this study, a multi-criteria decision making model is presented to determine higher priority waste types (air and solid wastes, wastewaters) among the three firms. We used stochastic data by applying and the SMAA-2 results are given. © 2015 Elsevier B.V., All rights reserved.
Citation - WoS: 12
Citation - Scopus: 13
Developing a Decision-Support System for Waste Management in Aluminum Production
(Springer, 2016) Ozmen, Mihrimah; Aydogan, Emel Kizilkaya; Ates, Nuray; Uzal, Nigmet; 01. Abdullah Gül University; 02.03. İnşaat Mühendisliği; 02. Mühendislik Fakültesi
Industrial enterprises constitute a major portion of the world's economy, as well as a large proportion of a country's businesses and total employment. In Turkey, industrial enterprises are underdeveloped in terms of knowledge, skill, capital, and particularly accessing and benefiting from the advantages provided by modern information and communication technologies. Aluminum manufacturing has been reported to be the largest industry in Turkey with respect to production volumes and application fields. However, aluminum production is known to be an important contributor to environmental pollution, and the relative contribution of other related enterprises to the total industrial environmental impact is unknown. Environmental pollution sources can typically be classified into three categories: gaseous emissions, solid wastes, and wastewaters. The types of wastes produced by aluminum production vary based on the process line used, the variety of target products produced, and the production capacity of a given plant. As the capacities of facilities grow, the type and amount of waste become more variable. Therefore, the primary objective of this study is to determine the priority of each waste type in aluminum manufacturing industries. This study was conducted in the Industrial Zone of Kayseri in Turkey. Three different facilities that range in size from large to small based on their production volume, plant capacity, and variety of production are selected for this study. The priority of waste types was determined by combining the AHP and PROMETHEE II multicriteria decision methods. While wastewater was categorized as having the highest priority in large facilities, solid waste was determined to be the highest priority in medium and small facilities.