Ayşe YILDIRIM - Academia.edu (original) (raw)

Papers by Ayşe YILDIRIM

Research paper thumbnail of An Improved Artificial Atom Algorithm with the Operator of Shuffled Frog Leaping Algorithm

Adıyaman Üniversitesi mühendislik bilimleri dergisi, Aug 31, 2022

Research paper thumbnail of Machine learning-based prediction of length of stay (LoS) in the neonatal intensive care unit using ensemble methods

Neural computing & applications, May 7, 2024

Neonatal medical data holds critical information within the healthcare industry, and it is import... more Neonatal medical data holds critical information within the healthcare industry, and it is important to analyze this data effectively. Machine learning algorithms offer powerful tools for extracting meaningful insights from the medical data of neonates and improving treatment processes. Knowing the length of hospital stay in advance is very important for managing hospital resources, healthcare personnel, and costs. Thus, this study aims to estimate the length of stay for infants treated in the Neonatal Intensive Care Unit (NICU) using machine learning algorithms. Our study conducted a two-class prediction for long and short-term lengths of stay utilizing a unique dataset. Adopting a hybrid approach called Classifier Fusion-LoS, the study involved two stages. In the initial stage, various classifiers were employed including classical models such as Logistic Regression, ExtraTrees, Random Forest, KNN, Support Vector Classifier, as well as ensemble models like AdaBoost, GradientBoosting, XGBoost, and CatBoost. Random Forest yielded the highest validation accuracy at 0.94. In the subsequent stage, the Voting Classifier-an ensemble method-was applied, resulting in accuracy increasing to 0.96. Our method outperformed existing studies in terms of accuracy, including both neonatal-specific length of stay prediction studies and other general length of stay prediction research. While the length of stay estimation offers insights into the potential suitability of the incubators in the NICUs, which are not universally available in every city, for patient admission, it plays a pivotal role in delineating the treatment protocols of patients. Additionally, the research provides crucial information to the hospital management for planning such as beds, equipment, personnel, and costs. Keywords Machine learning Á Ensemble methods Á Length of stay (LoS) prediction Á Neonatal intensive care unit (NICU) Á Classification & Ayse Erdogan Yildirim

Research paper thumbnail of Group elevator control optimization using artificial atom algorithm

2017 International Artificial Intelligence and Data Processing Symposium (IDAP), Sep 1, 2017

With the applications of artificial intelligence, it is possible to produce accurate and effectiv... more With the applications of artificial intelligence, it is possible to produce accurate and effective solutions for many problems encountered in daily life. In this study, Artificial Atom Algorithm, which is a topical algorithm, is used for optimum routing in group elevator control. The Artificial Atom Algorithm is a meta-heuristic algorithm that was improved by inspired the compounding process of atoms. With the application, it is aimed that is responded to hall calls in the most effective way by working 7 elevators in 16-storey building. The system which is designed for this purpose, offers optimum control for elevators. The total elapsed time to answer the hall calls has been minimized with the study.

Research paper thumbnail of Applications of artificial atom algorithm to small-scale traveling salesman problems

Soft Computing, Aug 2, 2017

Most of the meta-heuristic algorithms are based on the natural processes. They were inspired by p... more Most of the meta-heuristic algorithms are based on the natural processes. They were inspired by physical, biological, social, chemical, social-biological, biologicalgeography, music, and hybrid processes. In this paper, artificial atom algorithm which was inspired by one of natural processes was applied to traveling salesman problem. The obtained results have shown that for small-scale TSP, artificial atom algorithm is closer to optimum than the other compared heuristic algorithms such as tabu search, genetic algorithm, particle swarm optimization, ant colony optimization, and their different combinations.

Research paper thumbnail of Application of Three Bar Truss Problem among Engineering Design Optimization Problems using Artificial Atom Algorithm

2018 International Conference on Artificial Intelligence and Data Processing (IDAP), 2018

Optimization is all of the transactions made in order to search for the ideal. In the real world,... more Optimization is all of the transactions made in order to search for the ideal. In the real world, it is needed to optimization in many field. For this reason, optimization algorithms are popular topics that are frequently studied. In this study, it is focused on three bar truss problem which is one of the optimization problem in the engineering field. Artificial Atom Algorithm which is a new chemistry-based meta-heuristic optimization algorithm was used to solve this problem. The obtained results were compared with a Swarm Optimization Approach, Cuckoo Search Algorithm, Bat Algorithm, Mine Blast Algorithm and Cricket Algorithm which were used to solving the same problem.

Research paper thumbnail of Solutions of Travelling Salesman Problem Using Genetic Algorithm and Atom Algorithm

ABSTRACT Travelling Salesman Problem (TSP) is an optimization problem that aims navigating given ... more ABSTRACT Travelling Salesman Problem (TSP) is an optimization problem that aims navigating given a list of city in the shortest possible route and visits each city exactly once. When number of cities increases, solution of TSP with mathematical methods becomes almost impossible. Therefore it is better to use heuristic methods to solve the problem. In this study, for the solution of TSP, Genetic and Atom Algorithms which are based on heuristic techniques were used and their performances were compared. Genetic Algorithm which is an evolutionary algorithm is inspired by biological changes and it uses operators such as natural selection, reproduction, crossover and mutation. Population evolution is occurred by applying these operators iteratively and as the other heuristic techniques, it gives exact or approximate solutions. Another algorithm was applied for TSP is Atom Algorithm. This is a new meta-heuristic algorithm based on process of compound formation. To form a compound, there are two methods called as .onic Bond and Covalent Bond. Ionic Bond is formed by removing some electrons from some elements and attaching new electrons to these elements. In the other case, Covalent Bond is based the joint use of one electron by two elements. Change of atom set appears by applying these two operators iteratively. Atom algorithm differs from the other algorithms are in this field by taking into account the effect of parameter values on the solution. Covalent Bond operator is applied according to effect of electrons. In the applications of Genetic and Atom Algorithms for TSP, to avoid repetition of city, candidate solution sets were generated by codification in form of permutation and to prevent degradation of this form when operators were applied, some of the techniques were used. As a result, it was observed that for the solution of TSP, Atom Algorithm gives better results than Genetic Algorithm. That means, the salesman completes the tour by travelling less distance in Atom Algorithm. On the other hand, the results of Atom Algorithm had better stability. But Atom algorithm has a disadvantage in the aspect of time. It works slower than Genetic Algorithm.

Research paper thumbnail of Optimization of Fuel Cost in Electric Power Systems using Harmony Search Algorithm

Fuel Cost Optimization emerges as an important issue in electrical load distribution systems. In ... more Fuel Cost Optimization emerges as an important issue in electrical load distribution systems. In this study, the performance of Harmony Search Algorithm, which has a significant place in the literature, has been observed for sample problems in the field of fuel cost optimization. It is aimed to distribute the load provided by 3-unit, 13-unit and 40-unit power plants with minimum cost. Economic load dispatch problem emerges as a multi-objective optimization problem. For this reason, equal weighted scalarization method has been used. Transmission line losses are taken into account in the solution of unit-3 and unit-13 test systems. For this purpose, Kron's transmission line loss formula are used. In the solution of the unit-40 test system, transmission line loss is ignored. The results obtained are presented in comparison with various mathematical, evolutionary and heuristic/meta-heuristic algorithms used in the literature to solve the same problems. The results show that the Harmony Search Algorithm is a successful algorithm for fuel cost optimization in electric load dispatch systems.

Research paper thumbnail of Bireye Özgü Optimum Beslenme Çizelgesinin Yapay Atom Algoritması Kullanılarak Hazırlanması

Mustafa Kemal Üniversitesi Tıp Dergisi, 2015

Amaç: Beslenme, insanın sağlıklı bir yaşam sürdürebilmesinde önemli bir konudur. Bu nedenle, çalı... more Amaç: Beslenme, insanın sağlıklı bir yaşam sürdürebilmesinde önemli bir konudur. Bu nedenle, çalışmada, bireyin fiziksel, fizyolojik ve sosyolojik özellikleri göz önünde bulundurularak, günlük optimum beslenme çizelgesinin meta-sezgisel bir algoritma olan Yapay Atom Algoritması kullanılarak oluşturulması hedeflenmiştir. Gereç ve Yöntem: Çalışmada, kullanıcıdan cinsiyet, yaş, boy, kilo, hamilelik ve emzirme durumu ve dönemi, egzersiz durumu gibi bilgileri istenilerek bunlara göre kişinin günlük enerji gereksinimi belirlenmiştir. Başlangıçta rastgele oluşturulan günlük beslenme çizelgesi, bireyin günlük enerji gereksinimi ve besinlerin kalori değerleri göz önünde bulundurularak Yapay Atom Algoritmasıyla optimize edilmiştir. Bulgular: Kullanıcı bilgilerine uygun olarak günlük alınması gereken kalori miktarına en yakın değeri veren besin gruplarıyla bir beslenme çizelgesi elde edilmiştir. Rastgele seçilen 20 birey için Yapay Atom Algoritması ile elde edilen optimum beslenme çizelgelerinin ortalama standart hatası SEM (Standart Error of Mean) = 0.0314 olarak hesaplanmıştır. Sonuç: Böylece bireyin fiziksel, fizyolojik ve sosyolojik özellikleri göz önünde bulundurularak kişiye özgü beslenme planının meta-sezgisel bir algoritma olan Yapay Atom Algoritması yardımı ile rahatlıkla oluşturulabildiği görülmüştür.

Research paper thumbnail of An Improved Artificial Atom Algorithm with the Operator of Shuffled Frog Leaping Algorithm

Adıyaman Üniversitesi mühendislik bilimleri dergisi, Aug 31, 2022

Research paper thumbnail of Machine learning-based prediction of length of stay (LoS) in the neonatal intensive care unit using ensemble methods

Neural computing & applications, May 7, 2024

Neonatal medical data holds critical information within the healthcare industry, and it is import... more Neonatal medical data holds critical information within the healthcare industry, and it is important to analyze this data effectively. Machine learning algorithms offer powerful tools for extracting meaningful insights from the medical data of neonates and improving treatment processes. Knowing the length of hospital stay in advance is very important for managing hospital resources, healthcare personnel, and costs. Thus, this study aims to estimate the length of stay for infants treated in the Neonatal Intensive Care Unit (NICU) using machine learning algorithms. Our study conducted a two-class prediction for long and short-term lengths of stay utilizing a unique dataset. Adopting a hybrid approach called Classifier Fusion-LoS, the study involved two stages. In the initial stage, various classifiers were employed including classical models such as Logistic Regression, ExtraTrees, Random Forest, KNN, Support Vector Classifier, as well as ensemble models like AdaBoost, GradientBoosting, XGBoost, and CatBoost. Random Forest yielded the highest validation accuracy at 0.94. In the subsequent stage, the Voting Classifier-an ensemble method-was applied, resulting in accuracy increasing to 0.96. Our method outperformed existing studies in terms of accuracy, including both neonatal-specific length of stay prediction studies and other general length of stay prediction research. While the length of stay estimation offers insights into the potential suitability of the incubators in the NICUs, which are not universally available in every city, for patient admission, it plays a pivotal role in delineating the treatment protocols of patients. Additionally, the research provides crucial information to the hospital management for planning such as beds, equipment, personnel, and costs. Keywords Machine learning Á Ensemble methods Á Length of stay (LoS) prediction Á Neonatal intensive care unit (NICU) Á Classification & Ayse Erdogan Yildirim

Research paper thumbnail of Group elevator control optimization using artificial atom algorithm

2017 International Artificial Intelligence and Data Processing Symposium (IDAP), Sep 1, 2017

With the applications of artificial intelligence, it is possible to produce accurate and effectiv... more With the applications of artificial intelligence, it is possible to produce accurate and effective solutions for many problems encountered in daily life. In this study, Artificial Atom Algorithm, which is a topical algorithm, is used for optimum routing in group elevator control. The Artificial Atom Algorithm is a meta-heuristic algorithm that was improved by inspired the compounding process of atoms. With the application, it is aimed that is responded to hall calls in the most effective way by working 7 elevators in 16-storey building. The system which is designed for this purpose, offers optimum control for elevators. The total elapsed time to answer the hall calls has been minimized with the study.

Research paper thumbnail of Applications of artificial atom algorithm to small-scale traveling salesman problems

Soft Computing, Aug 2, 2017

Most of the meta-heuristic algorithms are based on the natural processes. They were inspired by p... more Most of the meta-heuristic algorithms are based on the natural processes. They were inspired by physical, biological, social, chemical, social-biological, biologicalgeography, music, and hybrid processes. In this paper, artificial atom algorithm which was inspired by one of natural processes was applied to traveling salesman problem. The obtained results have shown that for small-scale TSP, artificial atom algorithm is closer to optimum than the other compared heuristic algorithms such as tabu search, genetic algorithm, particle swarm optimization, ant colony optimization, and their different combinations.

Research paper thumbnail of Application of Three Bar Truss Problem among Engineering Design Optimization Problems using Artificial Atom Algorithm

2018 International Conference on Artificial Intelligence and Data Processing (IDAP), 2018

Optimization is all of the transactions made in order to search for the ideal. In the real world,... more Optimization is all of the transactions made in order to search for the ideal. In the real world, it is needed to optimization in many field. For this reason, optimization algorithms are popular topics that are frequently studied. In this study, it is focused on three bar truss problem which is one of the optimization problem in the engineering field. Artificial Atom Algorithm which is a new chemistry-based meta-heuristic optimization algorithm was used to solve this problem. The obtained results were compared with a Swarm Optimization Approach, Cuckoo Search Algorithm, Bat Algorithm, Mine Blast Algorithm and Cricket Algorithm which were used to solving the same problem.

Research paper thumbnail of Solutions of Travelling Salesman Problem Using Genetic Algorithm and Atom Algorithm

ABSTRACT Travelling Salesman Problem (TSP) is an optimization problem that aims navigating given ... more ABSTRACT Travelling Salesman Problem (TSP) is an optimization problem that aims navigating given a list of city in the shortest possible route and visits each city exactly once. When number of cities increases, solution of TSP with mathematical methods becomes almost impossible. Therefore it is better to use heuristic methods to solve the problem. In this study, for the solution of TSP, Genetic and Atom Algorithms which are based on heuristic techniques were used and their performances were compared. Genetic Algorithm which is an evolutionary algorithm is inspired by biological changes and it uses operators such as natural selection, reproduction, crossover and mutation. Population evolution is occurred by applying these operators iteratively and as the other heuristic techniques, it gives exact or approximate solutions. Another algorithm was applied for TSP is Atom Algorithm. This is a new meta-heuristic algorithm based on process of compound formation. To form a compound, there are two methods called as .onic Bond and Covalent Bond. Ionic Bond is formed by removing some electrons from some elements and attaching new electrons to these elements. In the other case, Covalent Bond is based the joint use of one electron by two elements. Change of atom set appears by applying these two operators iteratively. Atom algorithm differs from the other algorithms are in this field by taking into account the effect of parameter values on the solution. Covalent Bond operator is applied according to effect of electrons. In the applications of Genetic and Atom Algorithms for TSP, to avoid repetition of city, candidate solution sets were generated by codification in form of permutation and to prevent degradation of this form when operators were applied, some of the techniques were used. As a result, it was observed that for the solution of TSP, Atom Algorithm gives better results than Genetic Algorithm. That means, the salesman completes the tour by travelling less distance in Atom Algorithm. On the other hand, the results of Atom Algorithm had better stability. But Atom algorithm has a disadvantage in the aspect of time. It works slower than Genetic Algorithm.

Research paper thumbnail of Optimization of Fuel Cost in Electric Power Systems using Harmony Search Algorithm

Fuel Cost Optimization emerges as an important issue in electrical load distribution systems. In ... more Fuel Cost Optimization emerges as an important issue in electrical load distribution systems. In this study, the performance of Harmony Search Algorithm, which has a significant place in the literature, has been observed for sample problems in the field of fuel cost optimization. It is aimed to distribute the load provided by 3-unit, 13-unit and 40-unit power plants with minimum cost. Economic load dispatch problem emerges as a multi-objective optimization problem. For this reason, equal weighted scalarization method has been used. Transmission line losses are taken into account in the solution of unit-3 and unit-13 test systems. For this purpose, Kron's transmission line loss formula are used. In the solution of the unit-40 test system, transmission line loss is ignored. The results obtained are presented in comparison with various mathematical, evolutionary and heuristic/meta-heuristic algorithms used in the literature to solve the same problems. The results show that the Harmony Search Algorithm is a successful algorithm for fuel cost optimization in electric load dispatch systems.

Research paper thumbnail of Bireye Özgü Optimum Beslenme Çizelgesinin Yapay Atom Algoritması Kullanılarak Hazırlanması

Mustafa Kemal Üniversitesi Tıp Dergisi, 2015

Amaç: Beslenme, insanın sağlıklı bir yaşam sürdürebilmesinde önemli bir konudur. Bu nedenle, çalı... more Amaç: Beslenme, insanın sağlıklı bir yaşam sürdürebilmesinde önemli bir konudur. Bu nedenle, çalışmada, bireyin fiziksel, fizyolojik ve sosyolojik özellikleri göz önünde bulundurularak, günlük optimum beslenme çizelgesinin meta-sezgisel bir algoritma olan Yapay Atom Algoritması kullanılarak oluşturulması hedeflenmiştir. Gereç ve Yöntem: Çalışmada, kullanıcıdan cinsiyet, yaş, boy, kilo, hamilelik ve emzirme durumu ve dönemi, egzersiz durumu gibi bilgileri istenilerek bunlara göre kişinin günlük enerji gereksinimi belirlenmiştir. Başlangıçta rastgele oluşturulan günlük beslenme çizelgesi, bireyin günlük enerji gereksinimi ve besinlerin kalori değerleri göz önünde bulundurularak Yapay Atom Algoritmasıyla optimize edilmiştir. Bulgular: Kullanıcı bilgilerine uygun olarak günlük alınması gereken kalori miktarına en yakın değeri veren besin gruplarıyla bir beslenme çizelgesi elde edilmiştir. Rastgele seçilen 20 birey için Yapay Atom Algoritması ile elde edilen optimum beslenme çizelgelerinin ortalama standart hatası SEM (Standart Error of Mean) = 0.0314 olarak hesaplanmıştır. Sonuç: Böylece bireyin fiziksel, fizyolojik ve sosyolojik özellikleri göz önünde bulundurularak kişiye özgü beslenme planının meta-sezgisel bir algoritma olan Yapay Atom Algoritması yardımı ile rahatlıkla oluşturulabildiği görülmüştür.