IJET Vol. 9, Issue 3, September 2024
Bu koleksiyon için kalıcı URI
Güncel Gönderiler
Öğe The Impact of Irrationals on the Range of Arctan Activation Function for Deep Learning Models(İstanbul Gelişim Üniversitesi Yayınları / Istanbul Gelisim University Press, 2024) Tümer Sivri, Talya; Pervan Akman, Nergis; Berkol, AliDeep learning has been applied in numerous areas, significantly impacting applications that address real-life challenges. Its success across a wide range of domains is partly attributed to activation functions, which introduce non-linearity into neural networks, enabling them to effectively model complex relationships in data. Activation functions remain a key area of focus for artificial intelligence researchers aiming to enhance neural network performance. This paper comprehensively explains and compares various activation functions, particularly emphasizing the arc tangent and its specific variations. The primary focus is on evaluating the impact of these activation functions in two different contexts: a multiclass classification problem applied to the Reuters Newswire dataset and a time-series prediction problem involving the energy trade value of Türkiye. Experimental results demonstrate that variations of the arc tangent function, leveraging irrational numbers such as π (pi), the golden ratio (ϕ), Euler number (e), and a self-arctan formulation, yield promising outcomes. The findings suggest that different variations perform optimally for specific tasks: arctan ϕ achieves superior results in multiclass classification problems, while arctan e is more effective in time-series prediction challenges.Öğe Mathematical Optimization of Monte Carlo Simulation Parameters for Predicting Stock Prices(İstanbul Gelişim Üniversitesi Yayınları / Istanbul Gelisim University Press, 2024) Norozpour, SajedehStock price prediction holds paramount significance for individual investors, guiding crucial decisions in financial planning and investment strategies. This research delves into the methodology of Monte Carlo simulation, a versatile tool in financial modeling, to assess its advantages and disadvantages in the context of predicting stock prices. The study employs Python code to demonstrate the step-by-step implementation of Monte Carlo simulations, emphasizing the mathematical optimization of parameters for enhanced accuracy. Results showcase a characteristic bell curve, offering a probabilistic perspective on potential outcomes. Comparative analyses with other forecasting models, such as graphic analysis, underscore the superior reliability of Monte Carlo simulation in evaluating risks and rewards. Furthermore, the paper explores the application of Monte Carlo simulation in real-world scenarios, such as portfolio optimization and retirement planning, highlighting its pragmatic value for individual investors navigating the complexities of the stock market. The research concludes by acknowledging the limitations of the approach and advocating for a comprehensive consideration of all relevant factors in financial decision-making. This exploration serves as a valuable resource for individual investors seeking informed insights into probabilistic forecasting methods for effective stock price predictions.Öğe Shipping Containers as Temporary Shelters in Post-Disaster Scenarios: Flying Factories(İstanbul Gelişim Üniversitesi Yayınları / Istanbul Gelisim University Press, 2024) Yıldırım, Semih Göksel; Ciritci, İlke; Fındıkgil, Meryem Müzeyyen; Atalay, HilayAfter the recent earthquakes in Turkey, there has been a significant demand for temporary shelter. Issues such as the availability of emergency shelters, designated emergency assembly areas, and the lack of social networks have come to the forefront. Due to the construction industry's inability to produce the necessary quantity of prefabricated temporary housing, the Ministry of Trade imposed a three-month ban on the export of prefabricated structures in 2023. The limited availability of emergency assembly areas renders low-density temporary settlements unfeasible. For disaster victims, leaving their homes does not provide a solution to overcoming the trauma they have experienced; in fact, it can exacerbate other economic, social, and security issues. Reusable shipping containers can partially address the problem of temporary shelter by utilizing the concept of flying factories. This research proposes a model that encompasses both technical and social phases, including the creation of technical documentation prior to a disaster and aiming for a participatory production model in the aftermath. The establishment of temporary logistics and production facilities is crucial and should be driven by volunteer participation under the guidance of professionals. Additionally, this model includes training and coordination activities before a disaster as part of the execution plan. Through this study, which incorporates both physical and social dimensions, an integrated solution is proposed based on the identification of challenges faced after recent disasters.
