Bildiri Koleksiyonu | Enformasyon Teknolojileri Bölümü

Bu koleksiyon için kalıcı URI

https://hdl.handle.net/11729/5099

Listeleniyor 1 - 20 / 43

Yapay zekâ tabanlı yaklaşımlar ile canlı hayvan sayılarının bölgesel analizi ve öngörülmesi
(Liberty Publishing House, 2025-11-18) Aydın, Şahin; Karadağ, Yaşar; Seydoşoğlu, Seyithan
Hayvancılık, Türkiye tarım ekonomisinin temel taşlarından biri olup gıda güvenliği, kırsal kalkınma ve ulusal ekonomi açısından kritik bir rol oynamaktadır. Talep yapısındaki değişimler, üretim pratiklerindeki dönüşümler ve iklim kaynaklı belirsizlikler karşısında, canlı hayvan sayılarının güvenilir şekilde öngörülmesi her zamankinden daha önemli hale gelmiştir. Ancak geleneksel tahmin yöntemleri, hayvancılığın bölgesel farklılıklarını ve doğrusal olmayan dinamiklerini yakalamada çoğu zaman yetersiz kalmaktadır. Bu çalışmada, Türkiye’de canlı hayvan sayılarının bölgesel analizi ve geleceğe yönelik projeksiyonlarının oluşturulmasında yapay zekâ tabanlı yaklaşımların potansiyeli araştırılmaktadır. Çalışmada kullanılan veri seti, resmi istatistiklerden elde edilen ve uzun yılları kapsayan bölgesel canlı hayvan sayılarından oluşmaktadır. Öncelikle Exponential Smoothing ve ARIMA gibi klasik zaman serisi modelleri temel senaryo olarak uygulanmış, ardından elde edilen sonuçlar yapay zekâ tabanlı yöntemlerle karşılaştırılmıştır. Random Forest algoritması ile doğrusal olmayan ilişkilerin yakalanması, LSTM (Long Short-Term Memory) modeli ile de zaman serilerindeki uzun dönem bağımlılıkların öğrenilmesi hedeflenmiştir. Tahmin performansı RMSE ve MAPE hata metrikleriyle değerlendirilmiştir. Bulgular, Türkiye’de canlı hayvan sayılarının bölgeler arasında belirgin farklılıklar gösterdiğini ortaya koymaktadır. Batı bölgelerinde sürekli artış eğilimleri gözlemlenirken, bazı bölgelerde durağanlık öne çıkmıştır. Yapay zekâ tabanlı modeller, kısa, orta ve uzun vadeli öngörülerde geleneksel yöntemlere kıyasla daha yüksek doğruluk sağlamıştır. Çalışma, yapay zekânın öngörü doğruluğunu artırarak sürdürülebilir hayvancılık yönetimine ve kırsal kalkınmaya yönelik kanıta dayalı politika geliştirme süreçlerine katkı sağlayabileceğini göstermektedir.
Multi-product trend analysis, structural breaks, and 2030 projections of Türkiye’s agricultural imports: a policy-oriented evaluation
(Liberty Publishing House, 2025-10-28) Aydın, Şahin; Büyükaslan, Hasan; Delen, Veysel; Gürbüz, Hüseyin
Giriş ve Amaç: Türkiye’nin tarımsal üretiminde son yıllarda artan dışa bağımlılık, ithalat verileri üzerinden daha net ortaya konulabilmektedir. Bu çalışma, Türkiye’nin farklı tarımsal ürünlerdeki ithalat trendlerini incelemek, kriz yıllarının etkilerini değerlendirmek, 2030 yılına kadar projeksiyonlar yapmak ve politika perspektifinden sonuçlar üretmek amacıyla hazırlanmıştır. Böylece gıda güvenliği ve sürdürülebilir tarım politikalarına yönelik çıkarımlar sunulmaktadır. Gereç ve Yöntem: Çalışmada Türkiye’nin 2000–2024 dönemine ait resmi ithalat verileri kullanılmıştır. Veriler yıllık bazda derlenmiş, ürün türlerine göre sınıflandırılmış ve istatistiksel analizlere tabi tutulmuştur. Zaman serisi yöntemlerinden Holt-Winters üstel düzgünleştirme modeli ile 2025– 2030 dönemi için projeksiyonlar yapılmış; ayrıca kriz yılları (2008, 2009, 2018, 2020, 2022) ayrı olarak değerlendirilmiştir. Ürünler arası karşılaştırmalarda ortalama ithalat miktarı ve varyasyon katsayısı hesaplanmıştır. Bulgular: Analizler, özellikle buğday, mısır, soya ve ayçiçeği gibi stratejik ürünlerde ithalat bağımlılığının giderek arttığını göstermektedir. Kriz dönemlerinde (2008 küresel kriz, 2018 kur şoku, 2020 pandemi, 2022 savaş) ithalat miktarları belirgin şekilde yükselmiştir. Projeksiyonlara göre 2030 yılına gelindiğinde buğday ithalatının 14,5 milyon tona, ayçiçeği ithalatının 6,3 milyon tona, mısır ithalatının 4,6 milyon tona ve soya ithalatının 3,7 milyon tona ulaşması beklenmektedir. Tartışma ve Sonuç: Sonuçlar, Türkiye’nin tarımsal ithalatında artış eğiliminin devam edeceğini ve özellikle kriz dönemlerinde dışa bağımlılığın daha da görünür hale geldiğini ortaya koymaktadır. Bu durum, gıda güvenliği açısından risk teşkil etmektedir. Çalışma, yerli üretim kapasitesini artırmaya yönelik politikaların önemini vurgulamakta; stratejik stok yönetimi, tarımsal Ar-Ge yatırımları, üretim ve tedarik çeşitliliği gibi önerilerin altını çizmektedir.
Spatial distribution of Türkiye’s livestock products economy (1995–2020): sustainability-oriented visualization analysis
(Liberty Publishing House, 2025-10-25) Aydın, Şahin; Özkan, Oktay; Azgın, Şükrü Taner
Introduction and Purpose: Livestock production plays a strategic role in Türkiye’s agricultural economy and is directly linked to food security and sustainable development goals. The aim of this study is to examine the spatial distribution of the livestock products economy in Türkiye between 1995 and 2020, visualize regional differences, and reveal long-term trends. Materials and Methods: The study utilizes province-level annual livestock product values (in thousand TL) obtained from official statistical sources. The data were analyzed through spatial methods, including choropleth maps, trend analysis, and growth rate evaluations for selected crisis years (2001, 2008, 2018, 2020). The analyses were conducted from a sustainability perspective, and regional production centers were identified. Findings: The results show that the Marmara, Aegean, and Central Anatolia regions lead in livestock product values, while the Eastern and Southeastern Anatolia regions have recorded significant increases in recent years. Trend analysis indicates that Konya, İzmir, Erzurum, and Diyarbakır achieved the largest growth, whereas smaller provinces exhibited relatively limited increases. In terms of crisis years, the sector continued to grow except during the 2008 global financial crisis, with a notable increase observed during the 2020 pandemic. Discussion and Conclusion: Overall, Türkiye’s livestock products economy demonstrated a steady increase between 1995 and 2020. The findings suggest that while the sector is sensitive to global shocks, it remains relatively resilient to domestic crises and pandemic conditions. Spatial analyses highlight the necessity of considering regional disparities in the development of sustainable policies.
Regional analysis and forecasting of broiler and layer poultry production in Türkiye: a statistical and machine learning approach
(Liberty Publishing House, 2025-10-20) Aydın, Şahin; Gül, Osman Kubilay
Introduction and Purpose: As well as cattle farming and sheep & goat farming, poultry farming also has a significant place in Türkiye’s agricultural economy. There are two important branches, such as broiler and egg in this sector. There is not enough systematic research which examines the regional perspectives and provide future projections in poultry farming as in many areas of agriculture and livestock. The main purpose of this study is to analyze broiler and layer production in Türkiye, identify the main producing regions, and generate forecasts using both traditional statistical models and modern machine learning algorithms. Materials and Methods: The regional broiler and layer production datasets have been acquired from the web-based data platform of Turkish Statistical Institute (TÜİK). Top producer regions and long-term changes in broiler and layer chicken production have been identified using descriptive statistics. Two statistical techniques- Autoregressive Integrated Moving Average (ARIMA) and Exponential Smoothing (ES)- have been used to anticipate the total national production of broiler and egg chicken. Two machine learning models such as Random Forest and Gradient Boosting, nevertheless, have been created. Random Forest allows for assessing variable importance and capturing nonlinearities, and Gradient Boosting provides flexible parameterization (e.g., learning rate, tree depth) and can be tuned effectively to the dataset. The model performance has been evaluated by way of Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R². The projections for ten years have been generated. Results: The broiler chicken production has been largely concentrated on the north-west line. The top three producer regions are TR42 (Kocaeli–Sakarya–Düzce–Bolu–Yalova), TR33 (Manisa–Afyonkarahisar–Kütahya–Uşak), and TR22 (Balıkesir–Çanakkale) respectively. The models ES and ML envisioned moderate growth in broiler chicken production, on the other hand, the suggestion of ARIMA is a flatter trend. The top three producer regions in layer chicken production are TR33 (Manisa, Afyonkarahisar, Kütahya, Uşak), TR52 (Konya–Karaman), and TR83 (Samsun–Tokat–Çorum–Amasya) respectively. A slight decline from the recent peak has been indicated by ES. On the other hand, moderate growth has been referred to by ARIMA. ML models harmonized the differences between statistical models by drawing a more balanced growth path. Discussion and Conclusion: This research shows the importance of using both statistical and machine learning approaches together with the purpose of identifying the trend dynamics and nonlinear relationships in broiler and layer chicken production. The results reveal that north-western regions are leading in the broiler chicken production. On the other hand, western-central regions are dominating the layer chicken production. The results of this study can be utilized to create critical policy deductions and decisions of targeted investments by considering these distinct geographies. The proposed methodological framework can be adapted to other livestock production data as well.
IoT-based surveillance system for poultry farms using semantic web and deep learning
(Plusbase Akademi Publishing, 2023-04-30) Aydın, Şahin; Jones, Karl
Poultry diseases are among the most important problems encountered in poultry farming. Although detecting diseases before they infect all poultry seems to be a fundamental challenge, it is possible to detect poultry diseases in the early stages with an Internet of Things (IoT)-based surveillance system. IoT-based surveillance systems create an important opportunity to prevent the spread of diseases throughout the poultry house and to prevent the breeders from incurring financial losses. The internetbased surveillance system proposed within the scope of this study determines the presence of poultry and poultry species with the help of artificial intelligence (AI) and aims to prevent the spread of the disease to the entire poultry house by detecting diseased poultry with the data obtained from temperature sensors. The system will detect the presence of animals and the body temperature data of poultry in two different ways. The first is to detect the presence of the animal in the nests where the laying action is carried out using a weight sensor and to obtain the body temperature data of the poultry with a temperature sensor. Secondly, by using image processing techniques, it is to detect poultry roaming in the poultry house with a deep learning model and obtains body temperature data of poultry through infrared temperature sensors. The system will decide on the possible type of disease using ontologies related to poultry diseases according to the obtained body temperatures. As a result, this proposed system will enable early disease detection for poultry farms by using the perspectives of deep learning, semantic web, and ontology engineering disciplines, which are among the important fields of study in recent years.
Endüstri 4.0 bağlamında öğrenilmiş güçlülük ve karar verme ilişkisinde algılanan stres ve belirsizliğe tahammülsüzlüğün rolü
(Işık Üniversitesi Yayınları, 2022-05) Aktay, Habibe
Bir örgütün en temel amacı, ortalamanın üzerinde kar elde ederek sürdürülebilir rekabet avantajı kazanmasıdır. Teknolojik gelişimin hızına paralel olarak endüstriyel dönemler numaralandırılıp isimlendirilmiştir. Her bir dönem, bir teknolojik inovasyonun yaygınlaşması ile başlamıştır. Endüstri 1.0’da buhar gücü, Endüstri 2.0’da elektirik, Endüstri 3.0’da dijitalleşme ve Endüstri 4.0’da siber fiziksel sistemler öne çıkmaktadır. Siber fiziksel sistemlerin yaygınlaşması ile rekabetin koşulları tamamıyla değişmiştir. Bu teknolojik inovasyonlarla küreselleşme de hız kazanmış dolayısıyla örgütlerin sürdürülebilir rekabet avantajı kazanması gittikçe zorlaşmıştır. Endüstri 4.0’ın en ayırıcı özelliklerinden birisi değişim ve dönüşümlerin teknolojik ilerlemenin üssel büyüme doğasına uygun olarak çok hızlı gerçekleşmesi ve esnek üretim dolayısıyla belirsizliğin bu çağı domine etmesidir. Esnek üretimin artması ve üretimin globalleşmesi ile stresin daha yoğun olarak yaşanması kaçınılmaz hale gelmiştir. Öğrenilmiş güçlülük bireylerin stresle başa çıkma becerilerini arttıran ve yöneticileri engeller karşısında yılmayan bir pozisyona sokan beceriler toplamı olarak ifade edilmektedir. Örgütlerdeki insan kaynağının yeni dünya sisteminde etkin olabilmesi için, öğrenilmiş güçlülük düzeyinin yüksek olması ve stresi tolere edebilecek becerilere sahip olmasını gerekmektedir. Belirsizlik altında örgütlerin stratejik hedeflerine ulaşabilmeleri hem teknolojiyi amaçlara uygun kullanabilmeleri hem de örgütün en değerli sermayesi olan insan kaynağını etkili kullanabilmelerini gerektirmektedir. İnsan kaynağını etkin kullanmak öğrenilmiş güçlülük düzeyini yükseltip çalışanların daha rasyonel kararlar almaları ile mümkün olmaktadır. Bu çalışmanın amacı öğrenilmiş güçlülük ve karar verme ilişkisinde stres ve belirsizliğe tahammülsüzlüğün rolünü saptamaktır. Araştırmada nicel araştırma yöntemi ve tarama deseni kullanılmaktadır.
Edge detection of aerial images using artificial bee colony algorithm
(Selcuk University Faculty of Technology, 2021-11) Yelmenoğlu, Elif Deniz; Akhan Baykan, Nurdan; Taşdemir, Şakir
Edge detection techniques are the one of the best popular and significant implementation areas of the image processing. Moreover, image processing is very widely used in so many fields. Therefore, lots of methods are used in the development and the developed studies provide a variety of solutions to problems of computer vision systems. In many studies, metaheuristic algorithms have been used for obtaining better results. In this paper, aerial images are used for edge information extraction by using Artificial Bee Colony (ABC) Optimization Algorithm. Procedures were performed on gray scale aerial images which are taken from RADIUS/DARPA-IU Fort Hood database. Initially bee colony size was specified according to sizes of images. Then a threshold value was set for each image, which related with images’ standard deviation of gray scale values. After the bees were distributed, fitness values and probability values were computed according to gray scale value. While appropriate pixels were specified, the other ones were being abandoned and labeled as banned pixels therefore bees never located on these pixels again. So the edges were found without the need to examine all pixels in the image. Our improved method’s results are compared with other results found in the literature according to detection error and similarity calculations’. All the experimental results show that ABC can be used for obtaining edge information from images.
Impact of vaccines on the COVID-19 pandemic in Turkey
(2022-06-01) Yelmenoğlu, Elif Deniz; Elmas, Dilara
COVID-19 (coronavirus disease-2019 pandemic continues to threaten public health and this situation is raising great concern all over the world. With the development of different vaccines, it was aimed to end the epidemic and increase community immunity in the past years. The research reduced public anxiety but the extent of the impact of vaccines in the pandemic is should be under investigation. Because the degree of availability of the COVID-19 vaccines was differing both nationally and globally. This makes it important to investigate how effective vaccination is on the epidemic. The main aim of this study is to investigate the possible recovery impact of vaccination on the COVID-19 pandemic in Turkey. In addition, the rates of severe disease during the first 3 doses of vaccination were also examined in this study. The analyses are conducted based on Spearman, Kendall and Pearson's correlation by using the data of the Ministry of Health of the Republic of Turkey. The obtained results showed that there are strong correlations between vaccination and recovery.
İnsansız hava aracı ve Sentinel-2 görüntüleri kullanılarak ayçiçeği haritalamasına dayalı kovan yerleştirme karar destek sistemi
(BZT Turan Publishing House, 2025-12-31) Yelmenoğlu, Elif Deniz; Aydın, Şahin; Çavdaroğlu, Gülsüm Çiğdem; Deniz, Hüseyin; Pajenado, Rex S.; Dilli, Şirin
Ayçiçeği, yüksek nektar üretim kapasitesi nedeniyle gezici arıcılık faaliyetleri açısından stratejik öneme sahip tarımsal bitkilerden biridir. Ayçiçeği ekim alanlarının mekânsal ve zamansal dağılımı, arı kolonilerinin beslenme olanaklarını ve dolayısıyla bal verimini doğrudan etkilemektedir. Bu nedenle, arı kovanlarının uygun alanlara ve doğru zaman dilimlerinde yerleştirilmesi, gezici arıcılığın verimliliği açısından kritik bir karar sürecini oluşturmaktadır. Ancak mevcut uygulamalarda, kovan yer seçimi çoğunlukla arıcıların bireysel deneyimlerine ve sezgisel yaklaşımlarına dayalı olarak gerçekleştirilmekte; uzaktan algılama, görüntü işleme ve mekânsal analiz gibi veri temelli yöntemlerden yeterince yararlanılmamaktadır. Bu durum, potansiyel olarak verim kayıplarına ve kaynakların etkin kullanılmamasına yol açabilmektedir. Bu çalışmada, ayçiçeği yoğunluğunun doğru ve güvenilir biçimde belirlenmesi yoluyla kovan yerleştirme planlamasını desteklemeyi amaçlayan, çok ölçekli bir uzaktan algılama tabanlı karar destek çerçevesi önerilmektedir. Önerilen yaklaşım, saha ölçeğinde yüksek mekânsal çözünürlük sağlayan insansız hava aracı (İHA) görüntüleri ile bölgesel ölçekte geniş alan kapsama imkânı sunan Sentinel-2 uydu görüntülerinin entegrasyonuna dayanmaktadır. Çalışma alanı olarak, Türkiye’nin önemli ayçiçeği üretim merkezlerinden biri olan Kırklareli ili seçilmiş; veri seti, nektar üretiminin en yüksek olduğu ayçiçeği çiçeklenme dönemi dikkate alınarak oluşturulmuştur. Ayçiçeği tespiti, makine öğrenmesi tabanlı Random Forest sınıflandırma yöntemi kullanılarak gerçekleştirilmiş ve geliştirilen model %90,7 genel doğruluk değerine ulaşmıştır. Sınıf bazlı performans değerlendirmelerinde ise, ayçiçeği ekili alanlar ile ayçiçeği olmayan alanlar için F1-skoru her iki sınıf açısından da 0,91 olarak hesaplanmıştır. Bu sonuçlar, modelin hem nektar açısından zengin ayçiçeği alanlarını hem de ayçiçeği bulunmayan bölgeleri güçlü ve dengeli bir şekilde ayırt edebildiğini göstermektedir. Elde edilen ayçiçeği yoğunluk haritaları temel alınarak, ayçiçeği oranının yüksek olduğu alanlar arı kovanı yerleştirilmesi için uygun bölgeler olarak tanımlanmış; ayçiçeği yoğunluğunun düşük olduğu veya hiç bulunmadığı alanlar ise kovan yerleştirilmesine uygun olmayan bölgeler olarak değerlendirilmiştir. Çalışmadan elde edilen bulgular, çok ölçekli uzaktan algılama verilerinin makine öğrenmesi yöntemleriyle bütünleştirilmesinin, gezici arıcılık uygulamalarında veri temelli, güvenilir ve ölçeklenebilir karar destek sistemlerinin geliştirilmesine önemli katkılar sağlayabileceğini ortaya koymaktadır.
Hybridization strategies in swarm intelligence: the case of ABC–FA and ABC–RUN algorithms
(BZT Turan Publishing House, 2025-10-08) Yelmenoğlu, Elif Deniz; Pajenado, Rex S.; Dilli, Şirin
Metaheuristic optimization algorithms have become a very popular field of study in recent years due to their ability to effectively solve complex, multidimensional problems. In this study, the Artificial Bee Colony (ABC), Firefly Algorithm (FA), and Runge–Kutta (RUN) optimization algorithms, known for their good performance among metaheuristic methods, are compared with their hybrid variants ABC_RUN and ABC_FA. Five widely used benchmark functions were selected for performance evaluation, and the performance results of the algorithms were statistically evaluated using the Wilcoxon signed-rank test. Furthermore, convergence curves were generated to show the average performance of the algorithms, and average running times were calculated to examine the balance between accuracy and computational cost. The findings show that hybrid methods provide higher accuracy compared to classical methods, while the RUN algorithm has an advantage in terms of running time. This comparative analysis demonstrates that hybrid approaches can more effectively balance exploration and exploitation, increase global optimization performance, and are applicable to real-world problems.
A novel hybrid RUN-ABC optimization algorithm
(BZT Turan Publishing House, 2025-10-08) Yelmenoğlu, Elif Deniz; Pajenado, Rex S.; Dilli, Şirin
In recent years, with the development of technology, complex and high-dimensional problems have increased. The use of metaheuristic optimization algorithms in solving these complex problems has become an important research area. In this study, a new hybrid RUN-ABC optimization algorithm was developed by combining the RUN (Runge Kutta Optimization) algorithm and the ABC (Artificial Bee Colony) algorithm. By taking into account the powerful exploration capabilities of the ABC algorithm and the efficient exploitation capabilities of the RUN algorithm, the aim was to search for the best solution in a more balanced manner in the search space. Experiments were conducted on five different benchmark functions to evaluate the performance of the hybrid RUN-ABC method. In these experiments, the developed hybrid method ABC and RUN algorithms were compared based on the average best value, standard deviation, and convergence rate. Furthermore, the Wilcoxon signed-rank test (signrank) was applied to measure the performance between the algorithms. The results showed that the developed hybrid RUN-ABC algorithm outperformed both the RUN and ABC algorithms in most cases. The developed method demonstrated impressive performance in terms of achieving a global minimum and the stability of its results. This study demonstrates that the developed hybrid RUN-ABC method can be a powerful alternative and provides a basis for its future use in solving various complex problems.
A geospatial analysis of the parks, emergency assembly areas, and urban green spaces in Izmir districts
(IKSAD Publications, 2024-12-30) Çavdaroğlu, Gülsüm Çiğdem; Günay, Nazan
Ensuring equity in the allocation of public resources is a central objective for planners. In the context of planning, equitable distribution involves strategically placing resources or facilities to maximize accessibility for a diverse range of spatially distinct social groups. Equity in resource distribution has been a focal point of interest across numerous disciplines. The equity mapping method, which utilizes visualization techniques within geographic information systems (GIS), serves as a valuable tool for analyzing the spatial equity in the distribution of public resources. In this study, equity mapping was applied to parks, green spaces, and emergency assembly areas - resources of significant societal importance - to evaluate individual accessibility to these public facilities. The fundamental methodological approach to equity mapping involves overlaying the distribution of accessibility measures with socioeconomic data to analyze spatial variations in equity. This approach relies on spatial univariate, bivariate, or multivariate analysis, which examines the mapped data distributions and spatial patterns to identify and characterize spatial associations. The study answers eight research questions: (1) the number of emergency assembly areas per capita by district, (2) the number of children's playgrounds per capita among the population aged 0-19 by district, (3) the number of fitness areas per capita among the population aged 20 and above by district, (4) the amount of urban green space per capita among the entire population by district, (5) the distance of the nearest emergency assembly area to the district center, (6) the distance of the nearest emergency assembly area to the neighborhood center, (7) the number of parks within reach of the neighborhood center, (8) number of parks within 1 km of buildings on a district basis. Obtained fundamental patterns of inequity in the distribution of focused public resources in the study may help the municipalities better understand the current situation, make plans for the following years and ensure a more equitable distribution of public resources.
An analysis of the effects of external factors on Covid-19 projections
(ICONSOS Publishing House, 2021-05-10) Çavdaroğlu, Gülsüm Çiğdem; Nuhui, Agim; Yılmaz, Baha Ahmet
[No abstract available]
NFC loyal system on the cloud
(IEEE Computer Society, 2013) Coşkun, Vedat; Özdenizci Köse, Büşra; Ok, Kerem; Alsadi, Mohammed
FC (Near Field Communication) technology facilitates mobile phone usage of billions of people throughout the world that offers diverse services ranging from payment and loyalty applications to access keys for offices and houses. NFC technology is one promising technology that have adopted smart card as secure element (SE) to provide a secure area for the execution of multiple applications and storing sensitive data. In this paper, we developed a framework to integrate NFC loyal system onto the cloud. We adapted NFC mobiles together with their SE to the cloud (Infrastructure as a Service - IaaS) using Cloud Computing methodology.
Reviewing the effects of spatial features on price prediction for real estate market: Istanbul case
(IEEE, 2022-09-16) Ecevit, Mert İlhan; Erdem, Zeki; Dağ, Hasan
In the real estate market, spatial features play a crucial role in determining property appraisals and prices. When spatial features are considered, classification techniques have been rarely studied compared to regression, which is commonly used for price prediction. This study reviews spatial features' effects on predicting the house price ranges for real estate in Istanbul, Turkey, in the classification context. Spatial features are generated and extracted by geocoding the address information from the original data set. This geocoding and feature extraction is another challenge in this research. The experiments compare the performance of Decision Trees (DT), Random Forests (RF), and Logistic Regression (LR) classifier models on the data set with and without spatial features. The prediction models are evaluated based on classification metrics such as accuracy, precision, recall, and F1-Score. We additionally examine the ROC curve of each classifier. The test results show that the RF model outperforms the DT and LR models. It is observed that spatial features, when incorporated with non-spatial features, significantly improve the prediction performance of the models for the house price ranges. It is considered that the results can contribute to making decisions more accurately for the appraisal in the real estate industry.
Comparison of choreography vs orchestration based Saga patterns in microservices
(Institute of Electrical and Electronics Engineers Inc., 2022) Aydın, Şahin; Çebi, Cem Berke
Microservice Architecture (MSA) is a design and architecture pattern created to deal with the challenges of conventional software programs in terms of stream processing, highly available flexibility, and infrastructural agility. Despite the many advantages of MSA, designing isolated services using the autonomous Databases per Services paradigm is difficult. We realized that because each microservice will have its repository, ensuring data coherence between databases becomes difficult, especially in reversals, where operations transcend different sites. Distributed networked transactions and rollbacks can be efficiently handled using two-phase commitment methods in hardware virtualization using RDBMS databases. However, these approaches can't be used in micro-services with segregated NoSQL servers. Three issues have been addressed in this study: (i) investigate the implementation of event choreography and orchestration methods for the Saga pattern execution in MSA, (ii) existing reality suggestions on the saga pattern adoption and implementation besides the use cases, and (iii) introduce the disbursed transaction records and rollbacks challenges in isolated No-SQL databases with reliant collections in MSA.
Photogrammetric monitoring of an artificially generated landslide
(Copernicus GmbH, 2011-05-08) Akça, Mehmet Devrim; Gruen, Armin W.; Askarinejad, Amin; Springman, Sarah Marcella
According to pre-planned schedules, a series of two artificial rainfall events were applied to a forested slope in Ruedlingen, northern Switzerland. The experiments were conducted in autumn 2008 and spring 2009, the second of which resulted in mobilising about 130 m3 of debris. Both experiments were monitored by a photogrammetric camera network in order to quantify spatial and temporal changes. A 4-camera arrangement was used for the image acquisition. The cameras operated at a data acquisition rate of circa 8 frames per second (fps). Image measurements were made using the Least Squares image matching method, which was implemented in an in-house developed software package (BAAP) to compute 3D coordinates of the target points. The surface deformation was quantified by tracking the small (ping-pong and tennis) balls pegged into the ground. The average 3D point-positioning precision of ±1.6 cm was achieved in the first experiment and ±1.8 cm in the second experiment.
Realities & constraints of coal from energy and environmental perspectives
(University of Pittsburgh, 2010) Ekinci, Ekrem
[No abstract available]
Primitives of service oriented information system development
(2011) Aydın, Mehmet Nafiz; Yalçınkaya, Tayfun
One can find the long standing problems with conceptual modeling such as model transformation, reusability and agility in the classical information system development (ISD). New approaches may help in overcoming these issues in conventional ISD's problems. The service oriented approach which is one of the promising approaches, brings solutions to some existing problems. By adopting service orientation in ISD - which we call service oriented ISD, one can deal with the aforementioned problems. This study examines the very notion of service in the context of conceptual modeling method. In particular, we show how to support service modeling in terms of guidance (that is, the way of modeling). The proposed modeling support is demarcated along with a method at the foundational level. The paper is a conceptual and work-in-progress paper.
Usability of mobile voting with NFC technology
(Acta Press, 2010) Ok, Kerem; Coşkun, Vedat; Aydın, Mehmet Nafiz
Voting is a method to select one opinion or a person often following discussion, debate or an election campaign. After centuries of paper based voting ballots, electronic voting is used along with various technologies. One of the promising technologies is Near Field Communication (NFC) which allows data transfer between NFC-enabled devices and smart tags within a short distance. In this paper, we have presented a new type of a secure voting system, namely NFC voting, and evaluated the system's usability in a university council election with an executable prototype. Among other findings, we found that NFC voting satisfies electronic voting requirements and further increases the subjective usability of the proposed system.

Güncel Gönderiler