Yazılım Mühendisliği Bölümü

Bu koleksiyon için kalıcı URI

https://hdl.handle.net/20.500.13055/51

Listeleniyor 1 - 20 / 23

DeepMatch: A BERT-powered talent matchmaking approach
(Springer Nature Link, 2025) Gözüaçık, Necip; Topaloğlu, Atakan; Evren, Ayse Mine; Karakuş, Serkan; Akram Bennour; Ahmed Bouridane; Somaya Almaadeed; Bassem Bouaziz; Eran Edirisinghe
Consultancy companies aim to match their employees to customer projects based on their employee’s talents. Traditional matchmaking methodologies are founded on manual processes that rely on rules of thumb or algorithms that are based on handcrafted heuristics, which cause the matchings to be not only sub-optimal, but also time-consuming, subjective, and prone to human errors. In this paper, we propose a novel consultancy matching algorithm that utilizes BERT to semantically find the most optimal consultant-project matchings for a given set of consultants and projects, pairing relevant project specifications with consultant specifications using the JVSAP algorithm. In doing so, our proposed talent matchmaking system may be utilized to improve the accuracy and efficiency of consultancy matching, thereby facilitating more effective consultancy engagements. Our findings suggest that the pairings demonstrate a discernible alignment with human intuition, as evidenced by the consistent correlation between consultants possessing domain-specific expertise and projects characterized by corresponding thematic descriptions. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.
MelanoTech: Development of a mobile application infrastructure for melanoma cancer diagnosis based on artificial intelligence technologies
(IEEE, 2024) Tokatlı, Nazlı; Bilmez, Yakuphan; Göztepeli, Gürkan; Güler, Muhammed; Karan, Furkan; Altun, Halis
This preliminary work introduces MelanoTech, a mHealth application designed and implemented to offer a user-friendly and intuitive interface for the early diagnosis of melanoma, a kind of skin cancer with significant fatality rates [1]. The application demonstrates promising performance in segmentation and classification tasks by utilizing deep learning models with Generative Adversarial Networks (GANs) for data augmentation. MelanoTech achieves a comprehensive accuracy rate of 92%, with a segmentation model accuracy rate of 93% and a lesion detection accuracy rate of 90%. Finally, incorporating data augmentation approaches based on GANs resulted in a 5% enhancement in the model’s performance. These findings highlight the capacity of MelanoTech as a dependable tool for improving the early diagnosis of melanoma and decreasing the workload of physicians in Turkish public hospitals.
Log-Harmonic mappings associated with the sine function
(2024) Kumar, Sushil; Çetinkaya, Asena; Özkan Uçar, Hatice Esra
In this paper, we define new subclasses STlh(s) and CSTlh(s) of sine starlike log-harmonic mappings and sine close-to-starlike log-harmonic mappings, respectively, defined in the open unit disc D. We investigate representation theorem and integral representation theorem for functions in the class STlh(s). Further, we determine radius of starlikeness for functions in the classes STlh(s) and CSTlh(s).
A study on harmonic functions
(Bulgarian Academy of Sciences, 2024) Uçar, Mehmet Fatih; Özkan Uçar, Hatice Esra
The class of functions that have bounded boundary rotation and bounded radius rotation are the generalization of the convex and the starlike functions, respectively. The concept of such functions was introduced by L¨owner [1]. But he did not use the present terminology. It was Paatero [2,3] who systematically developed their properties and made an exhaustive study of the class of functions that have bounded boundary rotation. We will examine in this paper, the subclass of SH that is related to the class of functions that have bounded radius rotation.
Artificial intelligence-based fair allocation in NOMA technique: A review
(Bentham Science Publishers, 2024) Kırtay, Seda; Yıldız, Kazım; Böcekçi, Veysel Gökhan
Non-Orthogonal Multiple Access (NOMA) is an innovation that has great potential in wireless communication. It permits multiple users to efficiently allot a frequency band by adjusting their power allocations. Nevertheless, attaining fair power allocation in NOMA structures presents complex challenges that require specific models, extensive training data, and addressing issues of generalization. This review aims to explore the applications of Artificial Intelligence (AI) and Deep Learning (DL) methods to tackle the challenges associated with fair power allocation in NOMA systems. The focus is on developing strong AI-DL models and creative optimization methods specifically designed for dynamic environments to improve transparency and interpretability. This study explores a wide range of techniques, including Reinforcement Learning, Convolutional Neural Networks (CNN) for power allocation, Generative Adversarial Networks, Deep Reinforcement Learning, and Transfer Learning. The goal is to enhance various aspects, such as power allocation, user coupling, scheduling strategies, interference cancellation, user mobility, security, and deeplearning- based NOMA. Despite the difficulties, impartial power allocation algorithms based on AI and DL show promise in improving user performance and promoting fair power distribution in NOMA systems. This study emphasizes the significance of continuous research efforts to overcome current obstacles, enhance efficiency, and strengthen the dependability of wireless communication systems. This highlights the significance of NOMA as an advanced innovation for upcoming wireless generations that go beyond 5G. Future areas of study involve investigating federated learning and novel techniques for gathering data and utilizing interpretable AI-DL models to address existing constraints. Overall, this review highlights the potential of AI and DL techniques in achieving fair power distribution in NOMA systems. However, further investigation is crucial to addressing obstacles and fully exploring the capabilities of NOMA technology.
Transfer learning in severity classification in Alzheimer's : A benchmark comparative study on deep neural networks
(Altınbas University, 2024) Kırtay, Seda; Koçak, Muhammed Tayyip
Alzheimer's disease has become a condition of the brain that progresses over time and impacts a significant number of individuals worldwide. Early diagnosis, timely intervention and management of this disease process are very important in Alzheimer's disease. With regard to this study, we propose a transfer learning based early detection approach for Alzheimer's disease using Moderate Demented, Mild Demented, No Demented and Very Mild Demented classification sets. The proposed approach utilizes transfer learning based on the use of a deep neural network model that has been trained to extract features from brain imaging data. To evaluate the performance in transfer learning, a dataset of 6,400 images from brain MRI scans is augmented using data augmentation techniques and used in various convolutional neural network models the like VGG-19, Resnet-50, DenseNet-121, Inception-V3, VGG-16. The results are planned to show that these models achieve high sensitivity, specificity and high accuracy in detecting early signs of Alzheimer's disease. The study also emphasizes these advantages of using transfer methods of learning for early Alzheimer's detection by comparing it with various other deep learning models. The findings of this research suggest that transfer learning-based approaches can aid in the early detection of Alzheimer's disease., which affects millions of people, and offer a practical solution to classify cognitive impairment. With the proposed approach, it is shown that by helping clinicians to detect individuals at risk of Alzheimer's at an early stage, it will be possible to provide timely intervention and, in fact, better patient care. In terms of more effective applicability in clinical applications, the proposed approach can be applied to different and larger datasets and populations to make improvements and provide convenience to clinicians and patients. The best success rate of the models we used is achieved on the VGG19, RESNET50 KNN model with 99 percent.
Nuclei instance segmentation in colon histology images with YOLOv7
(Springer Nature, 2024) Yıldız, Serdar; Memiş, Abbas; Varlı, Songül
In histology image analysis, instance-based nuclei segmentation is one of the challenging tasks within the segmentation-guided studies since it is quite troublesome to detect each distinct nuclei instance of each nuclei type in images in contrast to the semantic segmentation in which all the image pixels of a nuclei type are labelled with the same mask ID although the segmented region may comprise of multiple instances. In this paper, an instance-based medical image segmentation task is addressed, and in this context, instances of multiple types of nuclei in colon histology images are aimed to be delineated distinctly. For the instance-based segmentation of the nuclei in colon histology images, the YOLOv7 algorithm and its built-in instance segmentation module are utilized. In the experimental studies performed on Colon Nuclei Identification and Counting (CoNIC) Challenge 2022 colon histology image dataset by using a 5-fold cross-validation performance evaluation strategy, nuclei instances belonging to 6 classes as the neutrophil, epithelial, lymphocyte, plasma, eosinophil and connective were segmented. To calculate the overall system accuracy, the quantification metrics of mean average precision (mAP) and mean panoptic quality (mPQ) were measured. In performance evaluations, quite promising accuracy values were obtained. The mAP values of 0.2885 and 0.2903, and mPQ values of 0.1659 and 0.1704 were observed by using the YOLOv7 algorithm. To the best of our knowledge, this is the first nuclei instance segmentation study with YOLOv7.
Eklemeli imalat yöntemiyle üretilen ABS cıvata numunesinin mekanik özelliklerinin incelenmesi
(Kahramanmaraş Sütçü İmam Üniversitesi, 2024) Koçak, Muhammed Tayyip; Bayraklılar, Mehmed Said; Ülkir, Osman
Bu çalışmada, eklemeli imalat (Eİ) yöntemiyle üretilen cıvata numunesinin boyutsal doğruluğu ve elastisite modülü hesaplanmıştır. Basılı cıvata örneğinin boyutu, etkin elastisite modülü ve bilgisayar destekli tasarım (CAD) modeliyle karşılaştırılarak tespit edildi. Bu model, katı modelleme yazılımı olan SolidWorks kullanılarak tasarlandı. Numunelerin üretiminde Eİ yöntemlerinden olan eriyik yığma modellemesi (EYM) kullanılmıştır. Bu yöntemde birçok termoplastik malzeme kullanılmakla birlikte, mevcut çalışmada cıvataların üretiminde akrilonitril bütadien stiren (ABS) tipi malzeme tercih edilmiştir. Gerçekleştirilen deneysel çalışmalar ile cıvata numunelerin tek eksenli çekme mukavemeti gözlemlenmiş ve gerilme-gerinim eğrileri kullanılarak esneklikleri tespit edilmiştir. Etkin elastisite modülü, ANSYS yazılımı kullanılarak bilgisayar simülasyonu ile sonlu elemanlar analizi yapılarak oluşturuldu. Gerçek zamanlı uygulanan çekme testi sonucunda en yüksek mukavemet değeri 35.258 MPa olarak ölçülmüştür. Deneysel çalışmalar neticesinde ölçülen çekme mukavemeti değerleri ile simülasyon sonuçlarının uyumlu olduğu tespit edilmiştir. Bu çalışma, gerçek dünya uygulamalarında kullanılmak üzere 3 boyutlu baskılı ABS cıvatalarının oluşturulması, test edilmesi ve optimize edilmesinin birçok yönünün anlaşılmasına yardımcı olacaktır.
Tıbbi görüntüler üzerinden kemik şekil yapılarının bilgisayar destekli analizi: Kalça eklemi üzerine araştırmalar
(Türkiye Klinikleri, 2023) Memiş, Abbas
İnsan iskelet sistemini oluşturan kemiklerin genel şekil hatlarının modellenmesi ve bu kemiklerde çeşitli faktörlere bağlı olarak meydana gelen şekil değişimlerinin sayısal analizi, klinik perspektiften oldukça kritik bir öneme sahiptir. Tıpta, özellikle de ortopedi ve travmatoloji disiplininde olmak üzere, çok çeşitli tıbbi görüntüleme cihazlarından alınan medikal görüntüler üzerinden vücut kemik bileşenleri görsel olarak incelenmekte ve yorumlanmaktadır. Günümüzde, mevcut bilgisayar teknolojisi ve güncel görüntü işleme tekniklerini içeren yazılım araçları, tıbbi görüntüler üzerinden kemiklerin ve diğer anatomik yapıların bilgisayar destekli şekil analizlerinin yapılmasına imkân vermekte ve bu sayede klinik değerlendirmelerde ciddi anlamda karar-destekleyici bir rol üstlenmektedir. Bu bölümde, tıbbi görüntüler üzerinden kemik şekil yapılarının bilgisayar destekli analizine yönelik bir inceleme çalışması sunulmuştur. Çalışma kapsamında özel olarak kalça eklemi ve proksimal femur üzerine yapılan şekil analiz araştırmalarına yoğunlaşılmış ve bu bağlamda, kemik şekillerini analiz etmek amacıyla kullanılan genel yöntemler, yaklaşımlar ve özgün teknikler ele alınmıştır.
Comparison of mechanical properties of samples fabricated by stereolithography and fused deposition modelling
(2023) Bayraklılar, Mehmed Said; Buldu, Abdulkadir; Koçak, Muhammed Tayyip; Ülkir, Osman; Kuncan, Melih
Additive manufacturing (AM) technology has attracted significant attention with the rapid fabrication of 3D parts for various applications. The two most popular techniques in this technology, Fused Deposition Modelling (FDM) and Stereolithography (SLA), make it possible to produce functional parts with complex shapes quickly and cheaply. Determining the mechanical properties of the parts fabricated by these methods is important in terms of efficient operation in the relevant fields. In this study, forty-five test specimens were fabricated using three different polymer materials (UVR, PLA, and ABS) in SLA and FDM type 3D printers, including tensile, compression, and 3-point bending tests. Samples are printed at a 75% fill rate according to ASTM standards. Experimental studies were carried out to determine the mechanical properties of the samples. Among the samples, the highest strength values in tensile, compression and bending test samples made of UVR material were 60.39 MPa, 127.74 MPa and 118.35 MPa, respectively. In addition to mechanical properties, hardness, and SEM analyses were performed to examine the surface roughness, surface topography, and composition of the samples.
Material selection for artificial femur bone using PROMETHEE-GAIA method
(American Society for Testing and Materials, 2024) Koçak, Muhammed Tayyip; Bayraklılar, Mehmed Said; Kuncan, Melih
When replacing bones and implants, choosing the right materials for the artificial bone and orthopedic implants is crucial to the procedure’s success. In this work, a thorough assessment of the literature was followed by a thorough and rigorous evaluation of prospective materials for prosthetic femurs using a multicriteria decision-making process known as PROMETHEE-GAIA (Preference Ranking Organization METHod for Enrichment Evaluation and Geometric Analysis for Interactive Assistance). The proposed approach was validated using a total of 12 assessment parameters, including density, tensile strength, and ultimate tensile strength, and 17 candidate materials. The significance of the chosen criteria is well described. These 17 candidate implant materials and the 12 assessment criteria were used to develop a choice matrix. Rankings over the prepared matrix were produced using the PROMETHEE-GAIA program. Ti-6Al-7Nb, Ti-6Al-4V, and ASTM F1537, Standard Specification for Wrought Cobalt-28Chromium-6Molybdenum Alloys for Surgical Implants (UNS R31537, UNS R31538, and UNS R31539), Co-Cr-W emerged as the top contenders and were demonstrated as possible materials for effective artificial femur materials because of the assessment. With a large number of pertinent criteria and a wide range of materials, this study offers a framework for the selection of implant materials. It also emphasizes how choosing materials carefully may increase the durability and efficiency of orthopedic implants.
Artificial intelligence in the education sector in Türkiye: Opportunities and challenges
(2023) Kırtay, Seda
The concept of artificial intelligence, which emerged with online education platforms and their applications in education, brings along platforms that offer students options such as interactive course materials by using artificial intelligence-supported learning methods. In this way, it allows not only learning but also measurement and evaluation tools to be used, and it can also provide feedback by measuring students' knowledge-skill levels. With these functions, the concept of Artificial Intelligence can determine learning management in education and offer personalized learning paths by determining student-specific learning methods. In this context, this article aims to provide information about Artificial Intelligence and Educational Applications used in Türkiye and to examine its advantages and disadvantages, especially in terms of how it affects students who have difficulty in accessing technology.
Mechanical shaft optimization: A study on static structural analysis and topological optimization in ansys
(2023) Koçak, Muhammed Tayyip; Bayraklılar, Mehmed Said
Shafts are extensively used in engineering fields, serving roles in power transmission and rotational movement, thus holding significant importance. This study focuses on analyzing the structure of a selected shaft model derived from research. Subsequently, topology optimization is applied based on the obtained findings. ANSYS software is utilized for performing analysis and optimization analysis. Following the completion of these analyses, the results are thoroughly examined. The optimization process resulted in a reduction of about 2.65% in the maximum stress and approximately 2.46% decrease in the maximum strain, indicating improved mechanical performance. However, an increase of about 33.24% in maximum deformation was observed, which warrants further consideration. Most notably, the weight of the shaft decreased significantly by approximately 57.81%, resulting in the creation of a much lighter model. These outcomes highlight the potential of topology optimization, demonstrating the ability to create lighter and stronger models while utilizing resources efficiently. Consequently, it becomes imperative to explore these outcomes further by modifying selected parameters to achieve optimal results and enhance the model's performance. This study successfully showcases the potential of topology optimization, paving the way for the creation of lighter and stronger models in engineering applications.
Briot–Bouquet subordination properties foranalytic functions Involvingchoi–Saigo–Srivastava integral operator
(Bulgarian Academy of Science, 2023) Özkan Uçar, Hatice Esra; Çetinkaya, Asena
We consider a new subclass of starlike functions involving the Choi–Saigo–Srivastava integral operator associated with the sine function in open unit disc. In view of this function class, we examine majorization properties and Briot–Bouquet differential subordination relations for such functions.
Healthcare service accessibility path planner: Unveiling a new era of intelligent appointment management systems based on outpatient prioritizing
(IEEE, 2023) Tokatlı, Nazlı; Koçak, Muhammed Tayyip; Kırtay, Seda; Göztepeli, Gürkan; Aktaş, İbrahim Serhat; Altun, Halis
In light of increased constraints on healthcare systems, particularly as a result of the pandemic, the importance of directing patients to the appropriate healthcare departments for individualized treatment based on their health conditions has been emphasized. Numerous healthcare institutions currently employ an online booking system that enables patients to schedule appointments. However, because patient requests are the main driving force behind this process, appointments with inappropriate departments or the bypassing of primary care facilities like general practice clinics frequently occur. Many studies proposed the use of AI-based chatbots and machine learning algorithms in healthcare systems to improve clinic operations, reduce patient wait times, and predict outpatient appointment no-show rates. This paper describes the conception and implementation steps of an innovative (mhealth app) that uses open AI tools to prioritize and classify outpatients based on their symptoms. Our AI-based appointment scheduling app will decide for the outpatient either to schedule appointments with primary care facilities or direct them to the appropriate healthcare department in hospitals only when absolutely necessary, thereby nurturing a more efficient, patient-centered healthcare service.
Histoloji görüntülerinde hücre çekirdeklerinin daha iyi bölütlenmesi için boyut-bazlı uyarlanabilir örnek budama
(IEEE, 2023) Yıldız, Serdar; Memiş, Abbas; Varlı, Songül
Bu bildiride, histoloji görüntülerinde yer alan hücre çekirdek örneklerinin daha başarılı bir biçimde bölütlenebilmesi amacıyla kullanılabilecek ve hücre tiplerine göre uyarlanabilir bir boyut-bazlı örnek budama yaklaşımı sunulmuştur. Gerçekleştirilen çalışmada, kolon histoloji görüntülerinde yer alan farklı hücre çekirdek tipleri öncelikle U-Net medikal görüntü bölütleme yöntemi ile anlamsal olarak bölütlenmiş ve her bir çekirdek tipi için bölüt haritaları oluşturulmuştur. Devamında, bölüt haritaları üzerinde Havza algoritması koşulmuş ve hücre çekirdek örnekleri elde edilmiştir. Son aşamada ise, uyarlanabilir boyut-bazlı örnek budama yaklaşımı ile, her bir çekirdek tipi için kabul edilebilir çekirdek boyutlarının altında kalan ve çekirdek örneği olarak tasnif edilmeyen örnek bölütler elenmiştir. CoNIC 2022 veri seti üzerinde gerçekleştirilen testlerde, uyarlanabilir boyut-bazlı örnek budama yaklaşımının, bu yaklaşımın kullanılmadığı normal bölütleme metodolojisine göre daha üstün başarı sağladığı gözlemlenmiş ve ortalama 0.5090 mPQ değeri ölçülmüştür.
Using artificial intelligence methods for detection of HCV-Caused diseases
(2023) Koçak, Muhammed Tayyip; Kaya, Yılmaz; Kuncan, Fatma
The Hepatitis C Virus (HCV) can cause chronic diseases and even lead to more serious conditions such as cirrhosis and fibrosis. Early detection of HCV infection is crucial to prevent these outcomes. However, in the early stages of infection, when symptoms are not yet evident, patients rarely undergo HCV testing. This highlights the need for alternative materials to guide HCV testing for early detection of the disease. In this study, we investigate the use of artificial intelligence technology to determine the disease status of individuals using blood data. A total of 615 individuals were included in the study. Preprocessing, filtering, feature selection, and classification processes were applied to the blood data. The correlation method was used for feature selection, where the features with high correlation values were selected and given as input to five different classification algorithms. The results of the study showed that the K-Nearest Neighbor (KNN) algorithm achieved the best classification success for detecting HCV patients, with a rate of 99.1%. This research demonstrates that artificial intelligence technology can be an effective tool for early detection of HCV-related diseases. The results indicate that the KNN algorithm can provide clear information about hepatitis infection from different blood values. Future studies can explore the use of other AI techniques and expand the sample size to improve the accuracy of the model.
Pedagogical classification of educational robots in pre-school teaching
(2023) Gümüş, Muhammed Murat; Kayhan, Osman; Korkmaz, Özgen; Altun, Halis; Yılmaz, Nihat
This study aims to create a rubric based on the pedagogical properties of educational robots for pre-school students and determine the compliance level with educational robot sets. In this sense, the study is considered a first and significant step toward selecting robots based on pedagogical-driven factors. For this aim, a mixed-method research design was employed. A qualitative method was used to create the rubric items, and the rubric development was also supported through a quantitative process by including expert opinions and ensuring content validity. Furthermore, a descriptive survey model, one of the quantitative designs, was used to examine the suitability of educational robots for the pre-school education level. As an outcome of this study, a rubric of four dimensions with 28 items related to the pedagogical features of educational robots in pre-school was created. Furthermore, widely used educational robots at the pre-school level, such as Kidoboto, Lego Wedo, Mbot, Lego Spike, Lego Ev3, and Matatalab, were evaluated by experts using the created rubric.
Digital health navigator: Preliminary work on a personal health assistant software for all health literacy level users in Turkey
(İzmir Ekonomi Üniversitesi, 2023) Tokatlı, Nazlı; Kömür, Fatma Nur; Koçak, Muhammed Tayyip; Kırtay, Seda; Göztepeli, Gürkan; Gül, Beyza; Altun, Halis
Today's digital health terminology is actually advanced med-ical technologies that include computer-assisted therapy, smartphone apps, and wearable technologies. These technologies offer significant po-tential for improving accees to immediate medical care, efficiency, clinical effectiveness, and personalization of many health problem therapies. In this paper. we will elobareate on the preliminary design steps of a per-sonalized health assistant application project (PHAS). The proposed ap-plication can be classified as a mobile health app and not a telemedicine application. The idea behind this application is to reduce the physicians' workload in hospitals while providing health care to the comm y with different health literacy levels by easily using the application when gen-eral assistance about any health issues or an overall health and wellness improvement is required.
Nuclei segmentation in colon histology images by using the deep CNNs: A U-Net based multi-class segmentation analysis
(Institute of Electrical and Electronics Engineers Inc., 2022) Yıldız, Serdar; Memiş, Abbas; Varlı, Songül
As is known, pathologists visually examine the tissue distributions by using microscopes traditionally. The rise in digital image processing and machine learning also allows high-performance computerized analysis of histology images taken with modern imaging systems. In general, histological image segmentation is the first step in the quantitative analysis of histology images. Therefore, a high-accuracy segmentation is essential for histology image analysis in most cases. In this paper, we performed a deep Convolutional Neural Networks (CNNs) based nuclei segmentation study on colon histology images. By using the U-Net biomedical image segmentation model, it is aimed to classify each pixel in colon histology images into one of the following 6 types of nucleus: epithelial, lymphocyte, plasma, eosinophil, neutrophil, connective tissue or classify it as the image background. In comprehensive experimental tests performed on Colon Nuclei Identification and Counting (CoNIC) Challenge dataset, commonly used segmentation and classification metrics were measured, and promising segmentation performances were achieved.

Güncel Gönderiler