Browsing by Author "Karabulut, Korhan"

Now showing 1 - 20 of 33

Citation - Scopus: 17
A Discrete Artificial Bee Colony Algorithm for the Permutation Flow Shop Scheduling Problem with Total Flowtime Criterion
(IEEE, 2010) M. Fatih Tasgetiren; Quan-Ke Pan; P. Nagaratnam Suganthan; Angela H-L Chen; Tasgetiren, M. Fatih; Suganthan, P. Nagaratnam; Karabulut, Korhan; Pan, Quan-Ke; Ince, Yavuz; Chen, Angela H.-L.; Wang, Ling
Very recently Jarboui et al. [1] (Computers & Operations Research 36 (2009) 2638-2646) and Tseng and Lin [2] (European Journal of Operational Research 198 (2009) 84-92) presented a novel estimation distribution algorithm (EDA) and a hybrid genetic local search (hGLS) algorithm for the permutation flowshop scheduling (PFSP) with the total flowtime (TFT) criterion respectively. Both algorithms generated excellent results thus improving all the best known solutions reported in the literature so far. However in this paper we present a discrete artificial bee colony (DABC) algorithm hybridized with an iterated greedy (IG) and iterated local search (ILS) algorithms embedded in a variable neighborhood search (VNS) procedure based on swap and insertion neighborhood structures. We also present a hybrid version of our previous discrete differential evolution (hDDE) algorithm employing the IG and VNS structure too. The performance of the DABC and hDDE is highly competitive to the EDA and hGLS algorithms in terms of both solution quality and CPU times. Ultimately 43 out of 60 best known solutions provided very recently by the EDA and hGLS algorithms are further improved by the DABC and hDDE algorithms with short-term search.
Citation - WoS: 5
Citation - Scopus: 11
A Discrete Artificial Bee Colony Algorithm for the Traveling Salesman Problem with Time Windows
(IEEE, 2012) Korhan Karabulut; M. Fatih Tasgetiren; Tasgetiren, M. Fatih; Karabulut, Korhan
This paper presents a discrete artificial bee colony algorithm (DABC) for solving the traveling salesman problem with time windows (TSPTW) in order to minimize the total travel cost of a given tour. TSPTW is a difficult optimization problem arising in both scheduling and logistic applications. The proposed DABC algorithm basically relies on the destruction and construction phases of iterated greedy algorithm to generate neighboring food sources in a framework of ABC algorithm. In addition it also relies on a classical 1-opt local search algorithm to further enhance the solution quality. The performance of the algorithm was tested on a benchmark set from the literature. Experimental results show that the proposed DABC algorithm is very competitive to or even better than the best performing algorithms from the literature.
Citation - WoS: 55
Citation - Scopus: 61
A hybrid iterated greedy algorithm for total tardiness minimization in permutation flowshops
(Elsevier Ltd, 2016) Korhan Karabulut; Karabulut, Korhan
The permutation flowshop scheduling problem is an NP-hard problem that has practical applications in production facilities and in other areas. An iterated greedy algorithm for solving the permutation flowshop scheduling problem with the objective of minimizing total tardiness is presented in this paper. The proposed iterated greedy algorithm uses a new formula for temperature calculation for acceptance criterion and the algorithm is hybridized with a random search algorithm to further enhance the solution quality. The performance of the proposed method is tested on a set of benchmark problems from the literature and is compared to three versions of the traditional iterated greedy algorithm using the same problem instances. Experimental results show that the proposed algorithm is superior in performance to the other three iterated greedy algorithm variants. Ultimately new best known solutions are obtained for 343 out of 540 problem instances. © 2017 Elsevier B.V. All rights reserved.
A new model for minimizing the electric vehicle battery capacity in electric\rtravelling salesman problem with time windows
(Tubitak Scientific & Technological Research Council Turkey, 2021) Kazım Erdoğdu; KORHAN KARABULUT; Erdoğdu, Kazım; Karabulut, Korhan
The growing pollution in the environment and the negative shift in the global climate compel authorities\rto take action to protect the environment and human health. Transportation is one of the major contributors to this\renvironmental decay. The harmful gases released to the air by the vehicles using petroleum fuel increase each day. One\rof the solutions is to make a gradual transition to electric vehicles. A major part of manufacturing an electric vehicle\ris to produce an efficient electric motor and battery for it. Reducing the manufacturing and operating costs of these\rcomponents will result in reducing the overall costs of electric vehicles. In this study a new variant of the electric\rtravelling salesman problem with time windows (E-TSPTW) was proposed. The objective function of the problem is to\rminimize the required initial battery capacity of the electric vehicle. For this goal a new energy consumption model\rconsidering the load of the vehicle was proposed with three scenarios. The proposed model was solved with a hybrid\rsimulated annealing algorithm for all these scenarios. The performance of the proposed method was compared to the\rsolutions found by a mixed integer linear programming model. The experimental results on the benchmark instances\rshow that up to a 35% reduction in initial battery capacity hence reduction in its cost is possible.\r
Citation - WoS: 3
Citation - Scopus: 3
A Populated Iterated Greedy Algorithm with Inver-Over Operator for Traveling Salesman Problem
(SPRINGER-VERLAG BERLIN, 2013) M. Fatih Tasgetiren; Ozge Buyukdagli; Damla Kizilay; Korhan Karabulut; Tasgetiren, M. Fatih; Kizilay, Damla; Buyukdagli, Ozge; Karabulut, Korhan; BK Panigrahi; PN Suganthan; S Das; SS Dash
In this study we propose a populated iterated greedy algorithm with an Inver-Over operator to solve the traveling salesman problem. The iterated greedy (IG) algorithm is mainly based on the central procedures of destruction and construction. The basic idea behind it is to remove some solution components from a current solution and reconstruct them in the partial solution to obtain the complete solution again. In this paper we apply this idea in a populated manner (IGP) to the traveling salesman problem (TSP). Since the destruction and construction procedure is computationally expensive we also propose an iteration jumping to an Inver-Over operator during the search process. We applied the proposed algorithm to the well-known 14 TSP instances from TSPLIB. The computational results show that the proposed algorithm is very competitive to the recent best performing algorithms from the literature.
Citation - WoS: 48
Citation - Scopus: 55
A variable iterated greedy algorithm for the traveling salesman problem with time windows
(ELSEVIER SCIENCE INC, 2014) Korhan Karabulut; M. Fatih Tasgetiren; Tasgetiren, M. Fatih; Karabulut, Korhan; Fatih Tasgetiren, M.
This paper presents a variable iterated greedy algorithm for solving the traveling salesman problem with time windows (TSPTW) to identify a tour minimizing the total travel cost or the makespan separately. The TSPTW has several practical applications in both production scheduling and logistic operations. The proposed algorithm basically relies on a greedy algorithm generating an increasing number of neighboring solutions through the use of the idea of neighborhood change in variable neighborhood search (VNS) algorithms. In other words neighboring solutions are generated by destructing a solution component and re-constructing the solution again with variable destruction sizes. In addition the proposed algorithm is hybridized with a VNS algorithm employing backward and forward 1_Opt local searches to further enhance the solution quality. The performance of the proposed algorithm was tested on several benchmark suites from the literature. Experimental results confirm that the proposed algorithm is either competitive to or even better than the best performing algorithms from the literature. Ultimately new best-known solutions are obtained for 38 out of 125 problem instances of the recently proposed benchmark suite whereas 15 out of 31 problem instances are also further improved for the makespan criterion. (C) 2014 Elsevier Inc. All rights reserved.
Akademik makaleler için yarı otomatik döküman sınıflandırma ve kod organizasyon sistemi
(2015) Öztürk, Alican; Albayrak, Raif Serkan; Karabulut, Korhan
In this thesis, the aim is to use the locally entered 'codes' (keywords in the document) to determine what the users' associated topic with that document corresponds to via WordNet's connections, synsets and hypernyms. WordNet has a neatly arranged structure that not only includes meaning for each sense of the word but also all the other words associated with it, in forms of hyponyms, hypernyms, synonyms, holonyms and meronyms. All of these words are connected in a network structure with appropriate links in between. By using the distance between the words to calculate the similarities between each pair of words inside a code cluster and enriching them with the hypernyms of high value nodes, it is possible to obtain a list of possible words that can be associated as topic keywords for the document itself. Since the codes entered into the system differ by the users' preferences and point of view on the document, it is highly possible for two instances to have completely different topics derived from the same document. The purpose of this is to personalize the topic according to the users' interest in the document instead of the presenting a generic topic about it. The project uses the Java library JWS to find the similarity between words and RitaWordNet from RitaCore to extract meanings and hypernyms of the words to select proper senses.
Citation - WoS: 31
Citation - Scopus: 33
An evolution strategy approach for the distributed blocking flowshop scheduling problem
(Elsevier Ltd, 2022) Korhan Karabulut; Damla Kizilay; M. Fatih Tasgetiren; Liang Gao; Levent Kandiller; Kizilay, Damla; Tasgetiren, M. Fatih; Gao, Liang; Karabulut, Korhan; Kandiller, Levent
Scheduling in distributed production environments has become common in recent years since the advantages of multi factory manufacturing have been growing. This paper examines the distributed blocking flowshop scheduling problem (DBFSP) to minimize the makespan. Two different mathematical models namely a mixed-integer programming model and a constraint programming model were proposed to solve the considered problem to optimality. Due to the NP-Hard nature of the problem large-size instances cannot be solved by the mathematical models and an evolutionary algorithm was proposed. Three different NEH-based heuristics were used and the first three solutions are included in the initial population whereas the rest is constructed randomly. The offspring population is generated by the self-adaptive destruction and construction (DC) procedure of the iterated greedy algorithm. Self-adaptive DC procedure is achieved by the evolution strategy approach. In the local search part of the algorithm a variable local search with three neighborhood structures was applied to the solution obtained by the DC procedure. The developed mathematical models initially verified the performance of the metaheuristic algorithm by using small instances. Then the proposed algorithm was tested on the benchmark suite from the literature. The computational results indicate that the proposed algorithm outperforms the other metaheuristic algorithms from the literature. Finally the solutions of the 156 best so far were obtained by the proposed algorithm which is more effective than the existing state-of-the-art methods. © 2021 Elsevier B.V. All rights reserved.
Citation - WoS: 42
Citation - Scopus: 45
An evolution strategy approach for the distributed permutation flowshop scheduling problem with sequence-dependent setup times
(Elsevier Ltd, 2022) Korhan Karabulut; Hande Oztop; Damla Kizilay; M. Fatih Tasgetiren; Levent Kandiller; Kizilay, Damla; Tasgetiren, M. Fatih; Karabulut, Korhan; Oztop, Hande; Kandiller, Levent
This paper considers a distributed permutation flowshop scheduling problem with sequence-dependent setup times (DPFSP-SDST) to minimize the maximum completion time among the factories. The global economy has enabled large companies to have distributed production centers to become widespread and effective production scheduling between these centers plays a vital role in the competitiveness of companies. To provide effective scheduling for the DPFSP-SDST we propose a new mixed-integer linear programming (MILP) model and a new constraint programming (CP) model which is presented for the first time in literature to the best of our knowledge. As the CP has become a solid competitor to the MILP in the literature this study aims to exploit the effectiveness of CP to solve such a complex DPFSP-SDST. Since the problem is NP-hard we also offer an evolution strategy (ES_en) algorithm that employs a self-adaptive scheme to obtain high-quality solutions in a short time. A ruin-and-recreate procedure is also embedded into the developed ES_en. We calibrate the parameters of the proposed ES_en using a design of experiment approach. We also compare the proposed ES_en algorithm's performance with three state-of-the-art metaheuristic algorithms from the literature i.e. the IG2S (a variant of an iterated greedy algorithm with NEH2_en initialization) IGR (another variant of an iterated greedy algorithm with a restart scheme) and discrete artificial bee colony (DABC) algorithm. A detailed computational experiment is carried out to evaluate the performance of the mathematical models (MILP and CP) and the heuristic algorithms (ES_en IG2S IGR and DABC). A comprehensive benchmark set is generated for the DPFSP-SDST from the well-known PFSP instances from the literature considering various combinations of jobs machines factories and SDST settings resulting in 2880 benchmark instances. For 216 out of 240 small-size instances optimal results are reported by solving the proposed MILP and CP models whereas time-limited model results are reported for the rest. The computational results show that the CP model outperforms the MILP model in terms of the solution time for small-size instances. Initially the performance of the heuristic algorithms is verified concerning the optimal results on small-size instances. Then the performance of the heuristic algorithms is evaluated for large instances. ES_en algorithm significantly outperforms the IG2S IGR and DABC algorithms for solving large instances. The computational results show that the proposed ES_en algorithm is robust and provides good-quality solutions for the DPFSP-SDST in a short computational time. © 2022 Elsevier B.V. All rights reserved.
Citation - WoS: 19
Citation - Scopus: 24
An evolution strategy approach to the team orienteering problem with time windows
(Elsevier Ltd, 2020) Korhan Karabulut; M. Fatih Tasgetiren; Tasgetiren, M. Fatih; Karabulut, Korhan
The team orienteering problem with time windows (TOPTW) is a highly constrained NP-hard problem having many practical applications in vehicle routing and production scheduling. The TOPTW is an extended variant of the Orienteering Problem (OP) where each node has a predefined time window during which the service has to be started. The aim is to maximize the total collected score by visiting a set of nodes with a limited number of tours since the given distance budget is limited. We propose an evolution strategy (ES) together with an effective constructive heuristic for solving the TOPTW. The main feature of the ES is to generate an offspring solution through ruin and recreate (RR) heuristic where a number of nodes are removed from the incumbent solution and then they are reinserted into tours until a complete solution is obtained. The ES is hybridized with an efficient random local search to enhance solution quality. For survivor selection we use a goodness of scores approach to determine and diversify the population for the next generation. Parameters of the ES are determined through the design of experiment approach to tune them. The computational results show that the constructive heuristic is slightly better than existing heuristics in the literature. Furthermore the detailed computation results on the benchmark suite from the literature confirm the effectiveness of the evolution strategy. Ultimately the evolution strategy obtains new best-known solutions for 7 benchmark problem instances. © 2019 Elsevier B.V. All rights reserved.
Android platformu için makine öğrenmesi teknikleri kullanarak kötücül yazılım tespiti
(2018) Peynirci, Gökçer; Eminağaoğlu, Mete; Karabulut, Korhan
Android mobil işletim sisteminin, rakiplerine kıyasla sahip olduğu oldukça yüksek toplam pazar payının yanında toplamda sayısal olarak çok daha fazla uygulamaya sahip olması dolayısıyla kötücül yazılımlar tarafından en sık hedef alınan mobil platform olduğu bilinmektedir. Son kullanıcının, tipik güvenlik yetersizliğine bağlı olarak, kötücül yazılımın Google Play Store veya herhangi bir resmi olmayan uygulama mağazasında yayımlanmadan önce tespit edilmesi hayati bir öneme sahiptir. Bu tezde, makine öğrenmesi teknikleri kullanarak yeni bir Android kötücül yazılım tespit metodolojisi yanında yeni bir öznitelik seçim metodolojisi ortaya konmuştur. Bu çalışmada sunulan makine öğrenmesi yaklaşımı, Android uygulamalarından (APK dosyaları) statik olarak çıkarılabilen, izinler (permissions), Uygulama Programlama Arayüzü çağrıları (API calls) ve katar (string) özelliklerini kullanmaktadır. Sunulan özellik seçim metodolojisinde literatürdeki mevcut yöntemlerden farklı olarak, belge sıklığı tabanlı (document frequency-based) bir yöntem tasarlanıp uygulanmıştır. Önerilen yöntem, Android kötücül yazılım örnekleri barındıran iki evrensel temel ölçüt veri kümesi ile test edilmiş ve bazı ikili sınıflandırma algoritmaları yanı sıra bazı topluluk (ensemble) yöntemine dayalı algoritmalar da kullanılarak literatürdeki diğer modeller ve yöntemlere göre daha başarılı sayılabilecek yüksek doğrulukta sonuçlar elde edilmiştir.
Artırılmış gerçeklik kullanan ev dekorasyon uygulaması
(2013) Doğan, Uğur Çağrı; Ak, Vedat Can; Karabulut, Korhan; Zincir, İbrahim
Artırılmış Gerçeklik, bulunduğumuz gerçek ortamı bilgisayar yardımı ile oluşturulmuş sanal nesneler ile birleştiren bir sistemdir. Sanal Gerçeklik, nesneleri tamamen sanal bir ortamda sergilediğinden, gerçek ortamda sanal nesneleri sergileyebilmek için Artırılmış Gerçeklik daha etkilidir. Bu tez, 3 boyutlu mobilya modellerini kullanıcının bulunduğu gerçek ortamda (özellikle kapalı ortamlarda), dekor değişikliklerini ön-izleme amacıyla görüntüleyen bir Artırılmış Gerçeklik uygulaması sunmaktadır. Tezin amacı, kullanıcılara, bulundukları gerçek ortamda 3 boyutlu mobilya modellerini ve ev tasarım eşyalarını göz önüne getirme imkânını verip, satın alma sürecinde kolaylık sağlamaktır. Bu uygulama, 3 boyutlu modellerin görselleştirmesi için üç teknik kullanmaktadır. İlki, modeli sahneye yerleştirebilmek için işaretleyici takibi yapan, işaretleyici tabanlı bir tekniktir. İkinci teknik de birincisi gibi işaretleyici tabanlı bir tekniktir; ancak birden fazla işaretleyiciyi takip eder. Her bir işaretleyici farklı mobilya modellerine atanmıştır. Ve sonuncusu, gerçek ortamdaki herhangi bir nesneyi, kullanıcıların takip işlemi için kullanabilmesini sağlayan işaretleyicisiz bir tekniktir. Ayrıca, bu tez SIFT ve eş düzlemli POSIT algoritmalarının sonuçlarını birleştiren bir bakış açısı sağlamaktadır. Önerilen sistem farklı ışıklandırma durumlarında başarıyla test edilmiştir.
Asimetrik Gezgin Satıcı Problemi İçin Bir Evrimsel Strateji Algoritması
(2016) KORHAN KARABULUT; Karabulut, Korhan
Gezgin satıcı problemi NP-zor sınıfındaki en bilinen problemlerden birisidir. En temel hali bile birçok pratik problemin modellenmesi için kullanılabildiği için üzerinde birçok akademik çalışma yapılmaktadır. Asimetrik gezgin satıcı problemi özellikle büyük şehirlerde ortaya çıkan tek yönlü yollar trafik sıkışıklığı gibi durumları da problem tanımına eklenebilmesine izin verir. Bu nedenle simetrinin her zaman geçerli olmadığı gerçek hayat problemlerinin modellenmesinde yaygın olarak kullanılmaktadır. Bu çalışmada asimetrik gezgin satıcı probleminin çözümü için bir evrimsel strateji algoritması önerilmektedir. Geliştirilen evrimsel strateji algoritması yeni çözümlerin üretilmesi için boz ve yeniden yap algoritmasını kullanmaktadır. Bu algoritmada belirlenen sayıda çözüm bileşeni rasgele seçilerek çözümden çıkartılır. Sonraki adımda çıkartılan bileşenler uygunluk değerini en küçük yapacak biçimde çözüme tekrar eklenir. Boz ve yeniden yap algoritması dolayısı ile evrimsel strateji algoritması için önemli bir parametre olan bozma boyutu evrimsel stratejisi algoritmasının özuyarlama özelliği kullanılarak güncellenmektedir. Özuyarlama özelliği algoritmanın iyi sonuç üreten parametre değerini öğrenmesini ve arama süresince aramanın o anki durumunun gerektirdiği biçimde değiştirilmesini sağlamaktadır. Elde edilen yeni çözümlerin daha da iyileştirilmesi için yerel arama aşamasında 3-opt algoritması kullanılmaktadır. Geliştirilen evrimsel strateji algoritmasının başarımının test edilmesi için literatürde en çok kullanılan problem kümesi olan TSPLIB kütüphanesi problemleri kullanılmış ve elde edilen sonuçlar sunulmuştur. Geliştirilen algoritma problemlerin optimum değerlerini çoğu zaman elde etmiş elde edemediği durumlarda da optimum değerden sapma en çok %1 olarak gerçekleşmiştir. Geliştirilen algoritmanın asimetrik gezgin satıcı probleminin kısa sürede etkin biçimde çözülmesi için kullanılabilecek bir yöntem olduğu gösterilmiştir.
Citation - WoS: 19
Citation - Scopus: 22
Bi-objective green vehicle routing problem
(WILEY, 2022) Kazim Erdogdu; Korhan Karabulut; Erdogdu, Kazim; Karabulut, Korhan
The green vehicle routing problem (GVRP) is a variant of the vehicle routing problem (VRP) which increasingly attracts many researchers in recent years due to the growing global environmental issues. As the transportation of the products grows the number of vehicles in fleets and the pollutants caused by these vehicles also grow which in turn negatively affects human health. In this paper a biobjective GVRP was studied. The two objectives are minimizing the total distance and minimizing the total fuel consumption of all vehicle routes. As a solution method an adaptive large neighborhood search was hybridized with two new local search heuristics. The proposed method was applied to two well-known benchmark problem sets for VRPs and new approximate Pareto fronts were obtained for these benchmark sets.
Bulut çizelgelme problemi için yeni bir güç tüketimi modeli
(2024) Kızıl, Alper; Karabulut, Korhan
Bulut bilişim, hesaplama gücü, grafik hesaplama gücü, depolama, bant genişliği, veri tabanı ve yazılım hizmetleri gibi bilgi işlem kaynaklarının dinamik olarak internet üzerinden kullanıcılara sunulmasıdır ve hem büyük hem de küçük şirketler ile geliştiriciler için maliyet tasarrufu, uygulamalarda esneklik ve ölçeklenebilirlik, kolay erişilebilirlik, güvenilirlik ve kolay afet kurtarma gibi pek çok avantaj sağlar. Öngörülebilir gelecekte, bulut bilişime olan talebin artacağı açıktır. Bulut veri merkezleri, tasarımlarının gereği olarak, önemli miktarda enerji tüketirler. Dolayısıyla, küçük tasarruflar bile daha büyük ölçekte önemli enerji tasarrufuna yol açabilir. Karbon nötr ve yeşil bilişimin giderek daha önemli hale gelmesiyle, bulut bilişimdeki en önemli sorunlardan biri olan ve NP Zor Problemi olduğu kanıtlanmış bulut kaynak planlaması, teorik olarak sınırsız sayıda kullanıcıya hizmet verebilecek sınırlı sayıda bulut kaynağı için en iyi çözümü bulmayı amaçlamaktadır. Bu çalışmada, farklı CPU mimarileri için güç tüketimi verilerini deneysel olarak toplanmış, bu deneysel verilerle yeni ve özgün bir güç modeli önerilmiştir. Ayrıca, Bulut Kaynak Planlama sorununda önemli iki metrik, toplam tamamlanma süresi ve güç tüketimi, farklı homojen ve heterojen veri merkezi senaryolarında farklı deterministik, sezgisel ve meta sezgisel tek amaçlı algoritmalar kullanılarak olası bir denge araştırılmıştır. Sonuçlar, belirli senaryolarda iki hedef arasında açık bir ödünleşim olduğunu göstermektedir. Bu senaryolar için, çok amaçlı ve tek amaçlı algoritmalar arasında bir karşılaştırma yapılmış ve ortak bir Pareto kümesi bulunmuştur.
Çok amaçlı bir yeşil araç rotalama probleminin çözümü için evrimsel algoritmalar
(2019) Erdoğdu, Kazım; Karabulut, Korhan
Green Vehicle Routing Problems (GVRPs) increasingly gain prominence due to the environmental issues created by the transportation vehicle fleets. The amount of CO2 emissions caused by the fossil fuel vehicles can be decreased by reducing the amount of fuel consumption of these vehicles. In this thesis, a Multi-Objective Green Vehicle Routing Problem (MOGVRP) was studied. Two objectives were taken into consideration in the problem: minimizing the total distance and minimizing the total fuel consumption of all vehicle routes. Two state-of-the-art methods NSGA-II and 𝝐-MOEA were adapted and applied for the solution of the problem, a multi-objective local search heuristic was proposed, and Path-Relinking heuristic was modified for the multi-objective problem.
Citation - WoS: 4
Citation - Scopus: 5
Distance and Energy Consumption Minimization in Electric Traveling Salesman Problem with Time Windows
(Institute of Electrical and Electronics Engineers Inc., 2020) Kazım Erdoǧdu; Korhan Karabulut; Erdogdu, Kazim; Karabulut, Korhan
As global pollution caused by transportation increases the need for cleaner energy becomes more significant each day. For this reason one of the recent global technological and scientific tendencies is to develop and include electric vehicles in transportation. In this paper an Electric Traveling Salesman Problem with Time Windows was studied by considering two objectives: minimizing the total distance and minimizing the total energy consumption. As a solution method the well-known Simulated Annealing algorithm was hybridized with a constructive heuristic and a local search heuristic. This algorithm was executed on a set of well-known benchmark instances from the literature separately for the two objectives and the results were presented. © 2020 Elsevier B.V. All rights reserved.
Citation - Scopus: 3
Effects of Parameters of an Island Model Parallel Genetic Algorithm for the Quadratic Assignment Problem
(Institute of Electrical and Electronics Engineers Inc., 2019) Alper Kizil; Korhan Karabulut; Kizil, Alper; Karabulut, Korhan
Quadratic Assignment Problem (QAP) is one of the most difficult combinatorial problems in the literature and has a diverse field of applications. This paper presents the results of experiments on the impact of parallelization of a sequential GA using island model. Both of the genetic algorithms are applied to the QAP. For the island model parallel GA we systematically change the number of islands and investigate the effects of dividing the same global population into a number of subpopulations. The number of islands is gradually increased to observe the effects on solution quality and speedup in total execution time using different problem instances. The results clearly indicate that while parallelized version outperforms sequential counterpart in both solution quality and total execution time an increasing number of subpopulations also positively effects the results until a critical point where every subpopulation has a certain number of individuals to be able to evolve independently. Beyond that point the performance of the algorithm begins to decrease. © 2020 Elsevier B.V. All rights reserved.
Citation - WoS: 8
Citation - Scopus: 14
Feature Selection for Malware Detection on the Android Platform Based on Differences of IDF Values
(SCIENCE PRESS, 2020) Gokcer Peynirci; Mete Eminagaoglu; Korhan Karabulut; Eminagaoglu, Mete; Peynirci, Gokcer; Karabulut, Korhan
Android is the mobile operating system most frequently targeted by malware in the smartphone ecosystem with a market share significantly higher than its competitors and a much larger total number of applications. Detection of malware before being published on official or unofficial application markets is critically important due to the typical end users' widespread security inadequacy. In this paper a novel feature selection method is proposed along with an Android malware detection approach. The feature selection method proposed in this study makes use of permissions API calls and strings as features which are statically extractable from the Android executables (APK files) and it can be used in a machine learning process with different algorithms to detect malware on the Android platform. A novel document frequencybased approach namely Delta IDF was designed and implemented for feature selection. Delta IDF was tested upon three universal benchmark datasets that contain Android malware samples and highly promising results were obtained by using several binary classification algorithms.
Genetik programlama ile hava kalitesi zamana bağlı seri tahminleme
(2018) Taşbaş, Su; Karabulut, Korhan
Bu çalışmada, genetik programlama kullanılarak hava kalitesi zamana bağlı seri tahminleme gerçekleştirilmiştir. Dünya Sağlık Örgütü ve diğer çevre ajanslarının raporlarına dayanarak, hava kirliliğinin neden olduğu ölümleri ve sağlık sorunlarını önlemek için hava kalitesi tahminlemenin ne kadar önemli olduğu gösterilmiştir. Bu çalışmanın temel amacı, hava kalitesi tahmini için genetik programlamanın kullanımını arttırmaya ve makine öğrenmesi yöntemleri ve otoregresif bütünleşmiş hareketli ortalama (ARIMA) ile yarışabilirliğini göstermeye katkıda bulunmaktır. Çalışmada bir yıl süreyle saatlik olarak ölçülmüş meteorolojik veriler kükürt dioksit ve parçacık madde gaz yoğunlaşmalarını tahmin etmek için kullanılmıştır. Zamana bağlı seri tahminleme problemi sembolik regresyon problemi olarak tanımlanmış ve Java tabanlı Evrimsel Hesaplama Araştırma sistemi (ECJ) kullanılmıştır. Tahminleme sonuçları genetik programlamanın performansını göstermek için çeşitli karar ağacı algoritmalarından ve ARIMA modelinden elde edilen sonuçlarla karşılaştırılmıştır. Karşılaştırmalar genetik programlamanın topluluk öğrenme yönteminden bile daha iyi performans sergilediğini göstermiştir.