Atılım Academic Archive (Atılım University)
Not a member yet
9347 research outputs found
Sort by
Leveraging Transformer Models for Enhanced Time Series Forecasting of Export Amounts
İhracat miktarlarının tahmin edilmesi, küçük ve orta ölçekli işletmelerin (KOBİ'ler) küresel pazarlarda rekabetçi kalabilmesi için çok önemlidir. Geleneksel makine öğrenimi yöntemleri, finansal verilerin her şirket için yıllık olarak kaydedildiği, düzensiz dalgalanmalar ve uzun vadeli bağımlılıklar sergileyen çok değişkenli çoklu zaman serisi analizinin karmaşıklıklarıyla başa çıkmakta genellikle zorluk yaşar. Bu zorlukların üstesinden gelmek için, yıllık tekrar eden finansal verileri kullanarak ihracat miktarlarını tahmin etmek amacıyla Transformatör tabanlı bir yaklaşım sunuyoruz. Gelişmiş dikkat mekanizmalarına sahip Transformatör modeli, her bir işletmenin dokuz yıllık verisini içeren veri setimizde Rastgele Orman (Random Forest) ve Uzun Kısa Dönemli Bellek (LSTM) modellerinden daha iyi performans göstermiştir. Veri setindeki zaman noktalarının sayısı azaltıldığında Transformatör modelinde önemli bir performans düşüşü gözlemlenmiştir. Bununla beraber, genişletilmiş bir zaman serisi kullanıldığında performansının önemli ölçüde artması, başarılı ve etkili sonuçlar elde etmek için yeterince uzun, özellik açısından zengin zaman serilerine ve etkili özellik mühendisliğine ihtiyaç duyulduğunu açıkça göstermiştir.Forecasting export amounts is crucial for small and medium-sized enterprises (SMEs) to remain competitive in global markets. Traditional machine learning methods often struggle with the complexities of multiple multivariate time-series analysis, where financial data is recorded annually for each company, showing irregular fluctuations and long-term dependencies. Address these challenges, we introduce a Transformer based approach for forecasting export amounts using annually repeated financial data. The Transformer model, with its advanced attention mechanisms, outperformed Random Forest and Long Short-Term Memory (LSTM) models on our dataset, which spans nine years for each enterprise. When the number of time points in the dataset was reduced, the Transformer model exhibited a significant drop in performance. However, its performance increased notably with the use of an extended time series, clearly showing that successful and impactful results require sufficiently long, feature rich time series, enhanced by effective feature engineering. These findings indicate that Transformer models can significantly improve the accuracy of forecasting complex time series based on financial data and offer valuable insights for SMEs and policymakers
Turkey's Foreign Policy Toward Iraq and Iran From the End of the Cold War To the Arab Spring
Bu tez, 1990-2010 yılları arasında Irak ve İran'a yönelik Türk Dış Politikasını incelemektedir. Çalışmanın konusu, Soğuk Savaş'ın sona ermesinden sonra Türkiye'nin komşu ülkeleri -Irak ve İran- için çok yönlü ve güvenlik dışı araçları uygulamasını ve Irak ve İran'a yönelik politikasının Türk dış politikasına yansımalarını incelemek ve analiz etmektir. Bununla birlikte, jeopolitik açıdan önemli bir noktada olan Türkiye'nin komşu ülkelerden algıladığı tehdit ve fırsatlara nasıl yanıt verdiği de tartışılacaktır. Aynı zamanda, Türkiye'nin 1990-2000 ve 2000-2010 yılları arasındaki Irak ve İran'a yönelik politikaları arasındaki farklılıklar ve bu farklılıklara yol açan birçok faktör çalışmada incelenecektir. Türkiye'nin 1990'larda Batı ile ittifak halinde hareket ettiği yönündeki literatürdeki bilgilerin aksine, Türkiye 1990'lardan bu yana özerklik aramıştır. Ancak, Türkiye'nin çabaları iç ve dış dinamikler nedeniyle sınırlı kalmıştır. Türkiye'nin özerklik arayışı 2000'li yıllarda artmış olsa da yaşadığı coğrafyaya ilişkin güvenlik ve risk algıları olan bir ülke olduğu için yeterli olmamıştır. Bu çalışmada, Türkiye'nin stratejik özerklik arayışının nispeten geç ve yetersiz kalmasının nedenleri Irak ve İran örnekleri üzerinden değerlendirilecektir. Çalışmada İran ve Irak'ın seçilmesinin nedenlerinden biri, iki ülkenin 1990'lı yıllardaki ABD karşıtı tutumlarının küresel siyasetin dışında kalmasına neden olması ve Türkiye'nin bu nedenle çok taraflı politikalar yürütememesidir. Tez, Soğuk Savaş sonrası ve 2000'li yılların ilk on yılında Türkiye'nin Irak ve İran'a yönelik yaklaşımına odaklanarak, özellikle diğer orta büyüklükteki liberal devletlerin artan özerklik arayışları içinde, 1990'lı yıllarda liberal bir orta büyüklükteki devlet olmasına rağmen Türkiye'nin sınırlı ve geç özerklik arayışının nedenlerini de incelemeyi amaçlamaktadır. Tez, Türkiye'nin Batılı savunma ve siyasi örgütlerin üyeleri arasında yer almasına karşın, bu iki komşusuyla ilişkilerinin, bölgenin barış havzasını koruma konusunda Batı'dan bağımsız, özerk kararlar alma iradesini ortaya koyduğunu ileri sürmektedir.This thesis examines the Turkish Foreign Policy towards Iraq and Iran between 1990 and 2010. This study examines and analyzes Turkey's application of multi-faceted and non-security tools to its neighboring countries -Iraq and Iran- after the end of the Cold War and the reflections of its policy towards Iraq and Iran on Turkish foreign policy. However, it will also discuss how Turkey, which is at a vital point in geopolitical terms, responds to the threats and opportunities it perceives from neighboring countries. At the same time, the differences between Turkey's policies towards Iraq and Iran between 1990-2000 and 2000-2010 and the many factors that led to these differences will be examined in the study. Contrary to the information in the literature that Turkey acted in alliance with the West in the 1990s, Turkey has sought autonomy since the 1990s. However, Turkey's efforts remain limited due to internal and external dynamics. Although Turkey's search for autonomy increased in the 2000s, it was not sufficient due to its being a country with security and risk perceptions regarding the geography in which it lives. Therefore, in this study, the reasons for Turkey's relatively late and inadequate search for strategic autonomy will be evaluated through the examples of Iraq and Iran. One of the reasons for choosing Iran and Iraq in the study is that the anti-US positions of the two countries in the 1990s caused them to be excluded from global politics and Turkey's inability to conduct multilateral policies for this reason. Through keeping the focus on Turkey's approach to Iraq and Iran after the Cold War and in the first decade of the 2000s, the thesis also aims to examine the reasons for Turkey's limited and late search for autonomy despite being a liberal medium-sized state in the 1990s, particularly in the increasing autonomy search of other medium-sized liberal states. The thesis argues that, although Turkey is among Western defense and political organizations' members, its relations with these two neighbors exhibit the will to make autonomous decisions independent of the West regarding protecting the region's peace basin
Dissident women's organizations as a counter-hegemonic actor in Turkey
Gündüz, Melisa/0000-0002-5196-944X; Gencoglu, Funda/0000-0001-8211-8624Could the Turkish women's movement, which has a strong reaction mechanism, be a constituent actor of counter-hegemony? The main reasons behind this question are the women's movement's deep-rooted history and its openness to combine theory with practice/action. When looked from the Chantal Mouffe and Ernesto Laclau's perspective of radical democracy theory, the women's movement appears to have a considerable potential of deciphering the existing hegemony and articulating the social demands which exclude and are excluded by the present-day hegemony in Turkey. This article tries to understand how women's movement in Turkey conceptualizes the existing power relations that constitute the neoliberal religio-conservative hegemony and how it responds to it.Funda GencogluAcknowledgementsThis article is based on Melisa Gunduz's MA thesis that was supervised by Funda Gencoglu
Real Options Valuation of Photovoltaic Investments: a Case From Turkey
Or, Bartu/0000-0001-8986-6713; Bilgin, Gozde/0000-0001-6899-453X; Birgonul, M. Talat/0000-0002-1638-2926Investments in renewable energy resources have become inevitable due to increasing energy demand and energy prices, diminishing non-renewable energy resources, and the outgrowth of carbon footprints. Photovoltaic (PV) systems offer high solar energy potential in sustainable energy production whereas their high initial costs necessitate critical strategic valuation of investments. Valuation with conventional methods has been challenging due to existence of uncertainties such as fluctuating PV panel prices, changing meteorological conditions with certain effects on power generation, and governmental policies on energy market regulations. This study aims to propose a real options approach to valuation of residential rooftop PV system investments considering these uncertainties and demonstrate benefits of this approach with an application on the residential PV investment decisions in Turkey. The proposed method, Real Options Valuation (ROV) with Least-Squares Monte Carlo Simulation (LSMC) considers the deferral option of the investor by utilizing stochastic simulations, the discounted cash flow method, linear regression, and backward dynamic programming and thus evaluates the effects of uncertainties on financial attractiveness of residential PV investments. The case study findings proved that ROV with LSMC having a 7-years deferral option supported the investment decision with realizable cost-effective options while "NPV method" resulted in an infeasible investment. Scenario analysis was also conducted to explore policy options that can be used to promote solar energy investments in Turkey. This study has a potential to have practical contributions for investors as well as implications for policy-makers.Scientific and Technological Research Council of Turkey (TUBITAK) [2210/A]; TUBITAKThis work was supported by a grant from the Scientific and Technological Research Council of Turkey (TUBITAK) [Subprogram code: 2210/A, General Domestic Graduate Scholarship Program] . TUBITAK's support is gratefully acknowledged
Why Do Some Followers Remain Silent in Response To Abusive Supervision? a System Justification Perspective
This study investigates how the system justification motive manifests in employees' voice/silence behavior at the workplace. It also explores the moderating effects of system justification on the linkage between abusive supervision and voice/silence behavior for blue- and white-collar employees. The field study generated responses from 905 employees in Turkey. Multi-group analysis reveals that the moderating effect of system justification motives varies by occupational class. In particular, the impact of abusive supervision on silence becomes more salient when white-collar employees endorse higher system justification motives. However, in the blue-collar sample, the absence of a moderating effect could be attributed to the strong main effect of system justification motives. The current study adds to the extant literature by applying a system justification perspective to voice and silence behavior by collar differences at work. It also provides important implications for managers in dealing with workplace mistreatment affecting all occupational groups, mainly when blue-collar employee silence is endemic and regulatory policies are inadequate.Social Science Citation Inde
Unveiling the Application Potential of Pbmo<sub>0.75</Sub>w<sub>0.25< Crystal: Linear and Nonlinear Optical Properties Through Ellipsometry
Isik, Mehmet/0000-0003-2119-8266PbMo0.75W0.25O4 compound is formed by replacing one quarter of the Mo atoms in the PbMoO4 with W atoms and has significant potential for optoelectronic applications. Optical properties of PbMo0.75W0.25O4 single crystal have been systematically investigated using ellipsometry measurements in the spectral range of 2.4-5.4 eV. The linear optical parameters, including refractive index, extinction coefficient, and absorption coefficient, were extracted from the obtained ellipsometry data. By analyzing spectral dependence of these parameters, band gap energy, critical point energy, and single effective oscillator parameters were determined. The refractive index spectrum was analyzed in the below band gap energy region by considering Cauchy and Sellmeier models. Additionally, nonlinear optical values were calculated, providing a comprehensive understanding of the optical properties of the PbMo0.75W0.25O4 single crystal. This study not only contributes to the fundamental understanding of the crystal's optical properties but also has potential implications for applications in optoelectronic devices and photovoltaics.Science Citation Index Expande
Optimizing Flexible Job Shop Scheduling Using the Adaptive Big-Bang Big Crunch Algorithm
Big Bang Big Crunch yöntemi, özellikle esnek iş atölyeleri ve makine ortamları bağlamında olmak üzere, karmaşık iş atölyesi çizelgeleme sorunlarını ele almak için umut verici bir yaklaşım sunmaktadır. Bu yöntem, üretim sistemlerinde makespan'ı, toplam gecikmeyi en aza indirmede ve iş yükü dengesini sağlamada etkinliğini kanıtlamıştır. Potansiyel çözümleri değerlendirmek için bir uygunluk fonksiyonu kullanarak ve bir dizi aşama boyunca nüfusun uygunluğunu artırarak iyileştirerek, Big Bang Big Crunch yöntemi, imalat işletmeleri için güçlü bir optimizasyon yaklaşımı sunmaktadır. Esnek iş atölyesi çizelgeleme bağlamında, Big Bang Big Crunch algoritmasını geleneksel iş atölyelerine kıyasla eklenen karmaşıklığı etkin bir şekilde ele alacak şekilde uyarlamak çok önemlidir. Bu uyarlama, algoritmanın esnek iş atölyeleri ve makine ortamları tarafından sunulan benzersiz zorluklara uygunluğunu sağlamak için değişiklikler ve geliştirmeler yapılmasını içerir. Ayrıca, pekiştirme öğrenimi, sinir ağları ve meta-sezgiseller gibi gelişmiş tekniklerin dahil edilmesi, çizelgeleme yöntemlerinin zekâsını ve dayanıklılığını önemli ölçüde artırabilir, bunların çeşitli dinamiklere u yum sağlamasını ve bireyselleştirilmiş özelleştirme gereksinimlerini karşılamasını sağlayabilir. Bununla birlikte, Big Bang Big Crunch algoritmasının performansı, problem örneği, problem büyüklüğü ve mevcut hesaplama kaynakları gibi faktörlerden etkilenebilir. Bu nedenle, algoritmanın uygulamasının dikkatli bir şekilde değerlendirilmesi ve kişiselleştirilmesi, optimal sonuçları elde etmek için temel oluşturur. Dahası, uygun kodlama ve kod çözme şemaları geliştirme, etkili çeşitlendirme ve yoğunlaştırma stratejileri ve uygunluk değerlendirme işlevleri ile ilgili zorlukları ele almak, Big Bang Big Crunch yöntemini esnek iş atölyesi çizelgelemesine uygularken istenen sonuçlara ulaşmak için çok önemlidir. Sonuç olarak, Big Bang Big Crunch yöntemi, üretim sistemlerinin verimliliğini artırmak ve imalat işletmelerindeki üretim sistemlerini optimize etmek için değerli bir potansiyel sunmaktadır. Bu çalışmada açıklanan önerilere uyarak, kuruluşlar bu yöntemi, esnek iş atölyesi çizelgelemesinin karmaşıklıklarını ele almak ve üretim sistemleri için zeki ve güçlü çözümler elde etmek için etkili bir şekilde kullanabilir.The Big Bang Big Crunch method presents a promising approach for addressing complex job shop scheduling problems, particularly in the context of flexible job shops and machine environments. This method has demonstrated effectiveness in minimizing makespan, total delay, and achieving workload balance in production systems. By utilizing a fitness function to evaluate potential solutions and iteratively improving the fitness of the population through a series of phases, the Big Bang Big Crunch method offers a robust optimization approach for manufacturing enterprises. the context of flexible job shop scheduling, it is crucial to adapt the Big Bang Big Crunch algorithm to effectively handle the added complexity compared to traditional job shops. This adaptation involves making modifications and enhancements to the algorithm to ensure its suitability for the unique challenges posed by flexible job shops and machine environments. Additionally, the incorporation of advanced techniques such as reinforcement learning, neural networks, and metaheuristics can significantly enhance the intelligence and robustness of scheduling methods, enabling them to adapt to various dynamics and meet individual customization requirements. However, the performance of the Big Bang Big Crunch algorithm may be influenced by factors such as the problem instance, the size of the problem, and the available computational resources. Therefore, careful assessment and tailoring of the algorithm's implementation are essential to ensure optimal results. Furthermore, addressing challenges related to developing proper encoding and decoding schemes, effective diversification and intensification strategies, and fitness evaluation functions is crucial for achieving the desired outcomes when applying the Big Bang Big Crunch method to flexible job shop scheduling. In conclusion, the Big Bang Big Crunch method offers valuable potential for optimizing scheduling processes and enhancing the efficiency of production systems in manufacturing enterprises. By considering the recommendations outlined in this study, organizations can effectively leverage this method to address the complexities of flexible job shop scheduling and achieve intelligent and robust solutions for their production systems
Revolution and Dictatorship. the Violent Origins of Durable Authoritarianism
[No Abstract Available
Hierarchical Classification of Analog and Digital Modulation Schemes Using Higher-Order Statistics and Support Vector Machines
Yalcinkaya, Bengisu/0000-0003-3644-0692Automatic modulation classification (AMC) algorithms are crucial for various military and commercial applications. There have been numerous AMC algorithms reported in the literature, most of which focus on synthetic signals with a limited number of modulation types having distinctive constellations. The efficient classification of high-order modulation schemes under real propagation effects using models with low complexity still remains difficult. In this paper, employing quadratic SVM, a feature-based hierarchical classification method is proposed to accurately classify especially higher-order modulation schemes and its performance is investigated using over the air (OTA) collected data. Statistical features, higher-order moments, and higher-order cumulants are utilized as features. Then, the performances of some well-known classifiers are evaluated, and the classifier presenting the best performance is employed in the proposed hierarchical classification model. An OTA dataset containing 17 analog and digital modulation schemes is used to assess the performance of the proposed classification model. With the proposed hierarchical classification algorithm, a significant improvement has been achieved, especially in higher-order modulation schemes. The overall accuracy with the proposed hierarchical structure is 96% after 5 dB signal-to-noise ratio value, approximately a 10% increase is achieved compared to the traditional classification algorithm.Science Citation Index Expande
A Comprehensive Study of the Effect of Scanning Strategy on In939 Fabricated by Powder Bed Fusion-Laser Beam
This study provides a comprehensive investigation into the effects of different scanning strategies on the material properties of IN939 fabricated using the PBF-LB process. The scanning strategies examined included alternating bi-directional scanning with rotation angles of 0 degrees, 45 degrees, 67 degrees, and 90 degrees between adjacent layers (named as shown), as well as alternating chessboard scanning with rotation angles of 67 degrees and 90 degrees (named as Q67 degrees and Q90 degrees). The results revealed that the 45 degrees and 67 degrees samples had the highest relative density, while the 0 degrees and Q67 degrees samples showed the highest average porosity. Moreover, various types of cracks, including solidification, solid-state, and oxide-induced cracks, were observed. Among the bi-directional scan samples, the 0 degrees sample displayed the most extensive cracking and the highest sigma max residual stress values in both XZ and XY planes. Conversely, the 45 degrees and 67 degrees samples exhibited fewer cracks. Notably, the lowest sigma max residual stress in the XZ planes among the bidirectional scan samples was observed in the 67 degrees sample. Additionally, microstructural analyses indicated differences in grain size and morphology, among the samples. Texture analysis indicated that the 0 degrees and 90 degrees samples exhibited strong cube textures, whereas the texture intensity weakened for the 45 degrees and 67 degrees samples. Moreover, the alternating chessboard scanning strategy led to rougher surfaces (higher Sa and Sz values) compared to the alternating bi-directional scanning strategy, regardless of the rotation angles. Furthermore, the microhardness values among the samples showed minimal variance, ranging between 321 + 14 HV and 356+ 7 HV.SFI and I-Form Advanced Manufacturing Centre [21/RC/10295_P2]; Pro-gramme for Research in Third Level Institutions (PRTLI) Cycle 5; European Regional Development Fund (ERDF) , part of the European Union; Metal Forming Center of Excellence at Atilim UniversityThis publication has emanated from research supported by a research grant from Science Foundation Ireland (SFI) under grant number 16/RC/3872 and is co-funded under the European Regional Development Fund. This publication has emanated from research sup-ported by the European Union's Horizon 2020 Research and Innovation Program under grant agreement No. 101138859 (DILAPRO) . This publication has emanated from research supported in part by a grant from SFI and I-Form Advanced Manufacturing Centre under Grant number 21/RC/10295_P2. For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. The optical microscope images were taken at the Nano Research Facility in Dublin City University which was funded under the Programme for Research in Third Level Institutions (PRTLI) Cycle 5. The PRTLI is co-funded through the European Regional Development Fund (ERDF) , part of the European Union Structural Funds Programme 2011-2015. The authors are grateful for EBSD characterizations support from the Metal Forming Center of Excellence at Atilim University. The authors also thank Middle East Technical University for the residual stress measurements.r publication has emanated from research supported in part by a grant from SFI and I-Form Advanced Manufacturing Centre under Grant number 21/RC/10295_P2. For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. The optical microscope images were taken at the Nano Research Facility in Dublin City University which was funded under the Pro-gramme for Research in Third Level Institutions (PRTLI) Cycle 5. The PRTLI is co-funded through the European Regional Development Fund (ERDF) , part of the European Union Structural Funds Programme 2011-2015. The authors are grateful for EBSD characterizations support from the Metal Forming Center of Excellence at Atilim University. The authors also thank Middle East Technical University for the re-sidual stress measurements.Science Citation Index Expande