{
    "exported_at": "2026-11-18 01:20:45",
    "count": 21,
    "records": [
        {
            "id": 216,
            "title": "A Comparative Benchmark of Fairness Metrics in Machine Learning",
            "title_ar": "",
            "title_en": "A Comparative Benchmark of Fairness Metrics in Machine Learning",
            "author": "Md Asif Shahariar",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2025,
            "language": "en",
            "research_domain": "responsible-ai",
            "research_subdomain": "xai",
            "categories": "machine-learning; responsible-ai; predictive-analytics; xai; fairness",
            "keywords": "",
            "abstract": "Machine learning is no longer limited to low-risk tasks, but it is also being used in critical decision-making systems. Consequently, machine learning has a direct effect on individuals' lives in domains such as recidivism prediction (COMPAS) and loan approval. Thus, errors or biases generated by such algorithms may have severe consequences in the real world. Therefore, our expectations of ML are not limited to accuracy. In addition to making correct predictions, we expect the models to behave in an unbiased manner toward individuals and groups. Many mathematical definitions of fairness have been proposed by researchers, but there have been few large, systematic comparisons of fairness metrics applied to standard machine learning models. The main task of this thesis is to examine in detail how the balance or trade-off between accuracy and fairness occurs in different models. The study used two socially sensitive datasets, COMPAS recidivism and Adult Income. These two datasets were used to train and test nine machine learning models. The models used in this study include nine widely used binary classifiers, ranging from simple and interpretable algorithms (e.g., Logistic Regression and Support Vector Machine) to high-performance ensemble methods (e.g., Random Forest, XGBoost, and LightGBM). Each model was carefully evaluated using Demographic Parity Difference (DPD) and Equalized Odds Difference (EOD) which are two important group fairness criteria, and traditional performance metrics including Accuracy and F1-score. The outcome of the experiment clearly shows a trade-off between accuracy and fairness despite the fact that the nature of this relationship varies greatly among the model architectures. Although ensemble algorithms including XGBoost and LightGBM achieved the highest predictive accuracy, they also had the largest fairness gaps. On the other hand, less complex models like Decision Trees and Multi-layer Perceptrons tended to be more fair and less accurate. One of the key findings in the hyperparameter sensitivity analysis is the fact that fairness is not fixed. It varied with hyperparameters for models like Logistic Regression and SVM, but was less responsive for others. Furthermore, statistical significance testing highlighted that while performance advantages were often robust, differences in fairness rankings were frequently statistically insignificant, cautioning against over-reliance on minor metric variations. This thesis develops a robust empirical foundation for fairness assessment in machine learning through broad benchmarking with both sensitivity analysis and statistical testing. This study serves as a methodological guide and a source of empirical evidence for practitioners and researchers, providing a framework for evaluating fairness based on evidence-based insights. These results indicate that fairness is not universal or transferable across models or datasets, it must be evaluated contextually, with awareness of model behavior and tuning sensitivity.",
            "source_url": "https:\/\/openalex.org\/W7132936585",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/232398\/4\/ShahariarMdAsif.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7132936585",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:47",
            "updated_at": "2026-11-15 16:52:47"
        },
        {
            "id": 252,
            "title": "Benchmarking machine learning models at the edge",
            "title_ar": "",
            "title_en": "Benchmarking machine learning models at the edge",
            "author": "Md. Masud Haider",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2024,
            "language": "en",
            "research_domain": "computer-vision",
            "research_subdomain": "",
            "categories": "machine-learning; computer-vision",
            "keywords": "",
            "abstract": "Recently the use of resource-constrained edge devices to perform machine learning tasks has gained substantial research attention. The need to perform such tasks arises from the vast amounts of sensors, and IoT and smart devices that generate large swaths of data at an enormous rate. To process such data in a real-time manner, cloud computing has proven to be suboptimal. As an alternative, edge devices paired with ML models can execute demanding tasks in diverse environments. This study explored a range of machine learning models coupled with several low and high-end edge devices to test their suitability for deployment in an edge environment. Unlike previous works, this experiment takes into account the initial heavy lifting that machine learning models and devices go through before performing the actual task of classification or detection. To systematically benchmark and gather results, this study divides the ML pipeline into initial and execution phases while thoroughly measuring each phase’s time and resource usage. This type of rigorous performance metric analysis enables decision-making about which machine learning model and devices have better suitability for latency and speed. Moreover, to add a device in the suit of existing ones and replicate the benchmark workflow, this study has automated the benchmark with the Ansible automation tool. The thesis research presents its findings from different devices. The division of phases allows for a clear understanding of the bottlenecks experienced by the devices and their capabilities. Moreover, an underlying crucial detail such as the warmup was revealed by scrutinizing the performances of the devices from the very beginning. The thesis concludes with performance highlights and limitations of the benchmark, identifying possible future research directions that can be developed further.",
            "source_url": "https:\/\/openalex.org\/W7133090547",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/161905\/2\/HaiderMd.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7133090547",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:48",
            "updated_at": "2026-11-15 16:52:48"
        },
        {
            "id": 337,
            "title": "Incorporation of deep learning strategies to improve QOS in VXLAN-based networks",
            "title_ar": "",
            "title_en": "Incorporation of deep learning strategies to improve QOS in VXLAN-based networks",
            "author": "Kumarawadu KR Siriniwansa",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2025,
            "language": "en",
            "research_domain": "deep-learning",
            "research_subdomain": "cnn",
            "categories": "machine-learning; deep-learning; reinforcement-learning; responsible-ai; ai-applications; predictive-analytics; cnn; rnn; xai; ai-smart-cities",
            "keywords": "",
            "abstract": "This study examines how to improve Quality of Service (QoS) management in VXLAN-based overlay networks by integrating deep learning (DL) techniques. Due to issues like fluctuating workloads, complicated virtualized settings, and increased data traffic, traditional QoS mechanisms frequently fail to provide optimal performance metrics like packet loss, latency, and jitter in modern networks. Examining how DL models, such as CNNs, RNNs, autoencoders, and reinforcement learning, can forecast traffic patterns, identify abnormalities, and optimize resource allocation, the study reviews current QoS strategies in VXLAN. With a focus on situations like SDN, data centers, and NFV, it examines the possible advantages and intrinsic drawbacks of using DL for real-time QoS enhancement. The report also suggests future directions for real-world application while identifying present research limitations in areas like data availability, model explainability, and secure deployment. In order to improve efficiency, security, and user experience, work ultimately promotes intelligent, adaptive QoS frameworks driven by DL that can handle the complexity of large-scale, virtualized network infrastructures.",
            "source_url": "https:\/\/openalex.org\/W7132872346",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/231186\/2\/SiriniwansaKumarawaduKumuthuRandima.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7132872346",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:48",
            "updated_at": "2026-11-15 16:52:48"
        },
        {
            "id": 466,
            "title": "Privacy conscious computer vision",
            "title_ar": "",
            "title_en": "Privacy conscious computer vision",
            "author": "Jukka Yrjänäinen",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2020,
            "language": "en",
            "research_domain": "computer-vision",
            "research_subdomain": "object-detection",
            "categories": "machine-learning; deep-learning; computer-vision; responsible-ai; data-mining; object-detection; fairness",
            "keywords": "",
            "abstract": "This work describes the design and the implementation of the distributed computer vision solution for person tracking and counting in a public museum. The system consists of an edge device fleet that does send data to a cloud server for further analysis. The key design goal is the protection of the privacy of the persons being monitored. This is achieved by a system that does not send actual images to the server, instead detected persons are represented with a feature vector extracted from the images. Privacy sensitive image data is not stored or transmitted anywhere in the system. Device design consists of Raspberry Pi single-board computers equipped with a neural network acceleration hardware and a camera module. These devices are used to locate a person from the camera view with the object detection neural network. After object detection a re-identification neural network is applied to the found object to generate a feature vector representation. This vector is sent to the cloud server. Based on the feature vectors it is possible to associate detected people across multiple cameras and moments time. However, it is not possible to reconstruct the original image from the feature vector. The experiments and performance measurements with edge devices show that the simultaneous use of deep neural networks for object detection and feature generation using relatively low-cost hardware is feasible. The design recommendation based on the experiments is that the use of a dedicated HW accelerator for running all neural networks is preferred. The analysis also show that the variation of the accuracy and computational complexity of used neural networks offers a range of feasible performance trade-offs. The data analysis method in the server using only feature vector data for tracking and clustering is evaluated. The experiments with publicly available image dataset indicate that with the proposed approach it is possible to approximate person count with reasonable accuracy.",
            "source_url": "https:\/\/openalex.org\/W3120794583",
            "pdf_url": "",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W3120794583",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:49",
            "updated_at": "2026-11-15 16:52:49"
        },
        {
            "id": 505,
            "title": "Descriptive Complexity and Graph Neural Networks",
            "title_ar": "",
            "title_en": "",
            "author": "Matias Selin",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2025,
            "language": "fi",
            "research_domain": "deep-learning",
            "research_subdomain": "gnn",
            "categories": "deep-learning; gnn",
            "keywords": "",
            "abstract": "Koneoppimisen sisällä kehittynyt syväoppiminen on erityisesti viimeisen vuosikymmenen aikana noussut yhdeksi merkittävimmistä tämänhetkisistä tutkimusaiheista [68]. Suuri osa kehityksestä on kuitenkin perustunut kroonisen epäluotettavaan empiiriseen tutkimukseen samalla kun aiheen matemaattinen teoria on jäänyt pienemmälle huomiolle [40]. Erityistapauksessa, jossa käsiteltävänä on graafimuotoinen data, ovat graafineuroverkot [33, 31] saavuttaneet merkittävän suosion huolimatta siitä, että niiden teoreettiset perusominaisuudet ovat pitkään jääneet epämääräisiksi. Esimerkiksi näiden mallien ilmaisuvoimaa, eli niiden kykyä ilmaista eri funktioita syötetyllä datalla, on vasta hiljattain ymmärretty paremmin [11]. Yhdeksi lupaavaksi lähestymistavaksi tässä on osoittautunut kuvaileva vaativuusteoria, jossa tavoitteena on löytää näitä malleja karakterisoivia logiikoita [41]. Tässä tutkielmassa käsittelen kuvailevan vaativuusteorian soveltamista graafineuroverkkoihin. Luvun 1 johdannon jälkeen aloitan tutkielman varsinaisen sisällön luvussa 2 kattavalla kirjallisuuskatsauksella aiheen historiasta, minkä jälkeen kuvailen yksityiskohtaisesti luvussa 3 miten Ahvosen ja muiden [8] rekurrenttien liukuluvuilla laskevien graafineuroverkkojen karakterisaatio laskevalla modaalisubstituutiokalkyylillä toimii. Kuvailen myös, miten globaalin lukemisen lisääminen graafineuroverkkoihin voidaan simuloida logiikan puolella tätä vastaavalla globaalilla laskevalla modaalitimantilla. Luvussa 4 tarkastelen erästä tunnettua graafineuroverkkojen erotteluvoiman rajaa eli Weisfeilerin ja Lemanin algoritmia [86, 61] sekä sitä, millä keinoin sen yli voidaan päästä. Eri vaihtoehdoista tehokkain on Saton ja muiden [77] kehittämä noodirandomointi, jossa – hieman yksinkertaistettuna – jokaiselle noodille asetetaan yksilöllinen tunniste. Osoitan uutena tuloksena, että Ahvosen ja muiden [8] karakterisaatiotulosta voidaan helposti laajentaa noodirandomoituun tapaukseen sallimalla logiikassa propositiosymbolien kvantifiointi. Tutkielman lopuksi tutkin tätä uutta logiikkaa tarkemmin muun muassa vertailemalla sitä erääseen sen muunnokseen, joka poikkeaa siitä syntaktisesti vain hieman mutta osoittautuu muilta ominaisuuksiltaan varsin erilaiseksi. Tämän luvun tulokset perustuvat yhteistyöhön Antti Kuusiston kanssa.",
            "source_url": "https:\/\/openalex.org\/W7133053626",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/232168\/2\/SelinMatias.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7133053626",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:50",
            "updated_at": "2026-11-15 16:52:50"
        },
        {
            "id": 523,
            "title": "Deep Learning-Based Vision System for Intelligent Object Detection and Robotic Sorting Solution",
            "title_ar": "",
            "title_en": "Deep Learning-Based Vision System for Intelligent Object Detection and Robotic Sorting Solution",
            "author": "Anastasiia Stepanova",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2026,
            "language": "en",
            "research_domain": "computer-vision",
            "research_subdomain": "object-detection",
            "categories": "machine-learning; deep-learning; computer-vision; robotics; ai-applications; object-detection; image-segmentation; ai-industry",
            "keywords": "",
            "abstract": "Manufacturing and production enterprises encounter digital development in the scope of Industry 4.0. Car workshops as part of the automotive industry aftermarket are in the transition from manual labour to automation and robotics solutions that aim to facilitate the workspace productivity by decreasing workers’ fatigue and errors. Automotive industry expects efficiency increase by 30% due to the deployment of autonomous robotics in both production and operation. Machine vision serving as the perceptive brain for automation systems allows robots to navigate in the unstructured environments of car workshops. Recent advancements in deep neural networks allow models to learn hierarchical features and patterns extracted from visual data and hence contribute to automated solutions in car workshops’ unstructured environments that present a certain challenge for machine learning. This thesis does not aim to replace human workers, but to provide aid in necessary but secondary task of the post-work tool sorting and arranging in a car service garage. In this project, a deep learning-based vision system was developed and evaluated. The primary task of the project included automatic identification, instance segmentation, and pose estimation of metallic wrenches in automotive workshop environment. Metallic tools represent a certain machine vision challenge due to the high specular reflection that causes significant segmentation mask fragmentation and affects performance of 3D cameras. This issue was addressed by utilizing a deeplearning-based keypoint detection as geometric anchors to interpolate and repair fragmented masks. The vision framework utilizes custom trained Detectron2 segmentation framework. The primary technical contribution of this thesis is the developed keypoint-guided mask reconstruction algorithm. Unlike standard frameworks that encounter fragmented segmentation masks on reflective objects, this system exploits detected semantic keypoints as stable geometric anchors that are further utilized to interpolate the mask and repair incomplete visual data. In this system, Zivid2 M70 industrial 3D camera was used for perception task to ensure high precision of depth values. The developed vision system was experimentally validated through three testing cases: static repeatability, rotational accuracy, and robustness under occlusion. The test results have shown high system stability: angular standard deviation of 0.17° and depth variance of 0.02 mm. The system achieved MAE of 1.33° and RMSE of 1.69° in a 180° rotational sweep. Furthermore, the developed algorithm to reconstruct fragmented masks has shown precise orientation calculation in occluded scenes with angular variance of 0.08°. Therefore, the proposed vision system provides accurate and reliable data for robotic manipulation and automated tool arranging tasks under unstructured conditions of car workshops.",
            "source_url": "https:\/\/openalex.org\/W7164869336",
            "pdf_url": "",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7164869336",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:50",
            "updated_at": "2026-11-15 16:52:50"
        },
        {
            "id": 596,
            "title": "Non-Markovian Modelling of Limit Order Books with Generative Adversarial Networks",
            "title_ar": "",
            "title_en": "Non-Markovian Modelling of Limit Order Books with Generative Adversarial Networks",
            "author": "Navindu Subuddhi Nawurunna Palliyaguruge",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2025,
            "language": "en",
            "research_domain": "generative-ai",
            "research_subdomain": "gan",
            "categories": "machine-learning; generative-ai; predictive-analytics; gan",
            "keywords": "",
            "abstract": "This thesis studies the use of Wasserstein GAN with Gradient Penalty and Temporal Attention augmented Bilinear Layers (TABL) to generate synthetic limit order book (LOB) data. The primary goal is to assess whether a non-Markovian conditioning of future snapshots on LOB history improves realism and mid-price movement prediction relative to a Markovian baseline. The proposed W-GAN models learn price dynamics implicitly from historical order flow, without relying on explicit mid-price movement labels. Results demonstrate that non-Markovian models consistently generate more realistic LOB snapshots and achieve higher F1 scores in mid-price movement classification than the baseline model. Among them, a history length n =30 performed the best in matching stylised facts and mid-price movements. They also incorporate LOB snapshots with empty queues in training, and learns to reproduce empty queues, but less frequently than in real data. Computational constraints required simplifying the model outputs, resulting in minor compromises to the quality of the generated samples. These findings highlight the benefits of non-Markovian conditioning in LOB generation and suggest promising directions for future research on model stability, scalability, and cross-asset generalisation.",
            "source_url": "https:\/\/openalex.org\/W7134871345",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/233187\/2\/NawurunnaPalliyagurugeNavindu.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7134871345",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:51",
            "updated_at": "2026-11-15 16:52:51"
        },
        {
            "id": 696,
            "title": "Privacy vs Accuracy: Trade-offs in Federated Learning",
            "title_ar": "",
            "title_en": "Privacy vs Accuracy: Trade-offs in Federated Learning",
            "author": "Muhammad Ahmed",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2025,
            "language": "en",
            "research_domain": "responsible-ai",
            "research_subdomain": "federated-learning",
            "categories": "responsible-ai; ai-applications; federated-learning; fairness; ai-cybersecurity",
            "keywords": "",
            "abstract": "In federated learning, model updates from distributed sources are vulnerable to inference attacks that can reconstruct private training data from gradient information.Traditional approaches to this problem apply differential privacy uniformly across all model parameters, creating a significant trade-off where stronger privacy guarantees substantially degrade model utility.This thesis investigates how differential privacy can be selectively applied to gradient components to enhance privacy protection while maintaining model performance.We introduce selective gradient perturbation, a percentile-based approach that applies noise only to the most significant gradient components by magnitude.Through comprehensive experiments on the CIFAR-10 dataset using a simulated federated learning environment with 10 clients, we demonstrate that selective perturbation fundamentally transforms the conventional privacy-utility trade-off.Our findings reveal that by applying noise to only the top 10% of gradients, we achieve 74.38% accuracy while maintaining privacy protection, and counter-intuitively, stronger privacy settings (ε = 0.1) yielded better model performance (78.60% accuracy) than weaker protection (ε = 0.5, 66.56% accuracy).These results challenge the conventional wisdom that privacy necessarily comes at a significant cost to utility, offering a practical approach for implementing differential privacy in federated learning that achieves both strong privacy guarantees and high model utility.",
            "source_url": "https:\/\/openalex.org\/W7132924904",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/226252\/2\/AhmedMuhammad.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7132924904",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:52",
            "updated_at": "2026-11-15 16:52:52"
        },
        {
            "id": 720,
            "title": "Machine Learning Algorithms for Computer-aided Diagnosis of Myocardial Infarction from 2D Echocardiography and COVID-19 from Chest X-rays",
            "title_ar": "",
            "title_en": "Machine Learning Algorithms for Computer-aided Diagnosis of Myocardial Infarction from 2D Echocardiography and COVID-19 from Chest X-rays",
            "author": "Aysen Degerli Ahishali",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2024,
            "language": "en",
            "research_domain": "medical-ai",
            "research_subdomain": "medical-imaging",
            "categories": "machine-learning; deep-learning; computer-vision; medical-ai; ai-applications; image-segmentation; medical-imaging; ai-cybersecurity",
            "keywords": "",
            "abstract": "The importance of e-health has drastically increased as the world has engaged sharply in the digital world, which provoked computer-aided diagnosis (CAD) algorithms to become a necessity for accurate and automatic detection and differential diagnosis of various diseases and medical conditions. Recent advancements in machine learning brought novel solutions for more complex clinical tasks that improved medical diagnostics. Accordingly, machine and deep learning methods are widely utilized in CAD-based techniques, which analyze medical data by performing pattern recognition to reveal any anomalies to guide medical doctors and experts in clinical tasks. However, in the development phase of CAD algorithms, there are several limitations and challenges regarding the size, quality, and availability of medical data for analysis using machine learning. Hence, there is an urgent need for accurate, reliable, and robust CAD algorithms to overcome the limitations and challenges in medical diagnostics. This thesis focuses on providing machine learning algorithms for CAD-based detection and assessment of myocardial infarction (MI) and coronavirus disease 2019 (COVID-19) using apical 4-chamber (A4C) and apical 2-chamber (A2C) view 2D echocardiography recordings and chest X-ray (CXR) images, respectively. The hypothesis defined in this thesis is that the proposed CAD algorithms using machine learning can provide assistive and visualization tools in clinical practice for the detection and assessment of MI and COVID-19. Accordingly, the thesis contributions are categorized into computer-aided diagnosis of MI and COVID-19, where task-specific solutions are developed for the assessments. The most severe manifestation of coronary artery disease is myocardial infarction, which is colloquially known as heart attack that leads to irreversible necrosis of the myocardium and even death. The detection of MI requires the assessment of regional wall motion abnormality of the left ventricle that can be primarily monitored and evaluated with 2D echocardiography. This thesis contributes to a pioneer 2D echocardiography dataset, HMC-QU for the detection of MI over single-view and multi-view echocardiography. Accordingly, another contribution of this thesis is a generic pseudo-labeling annotation approach that accelerates and enhances pixel-level manual annotations by preventing the subjective and time-consuming annotation process. In single-view and multi-view echocardiography assessments, this thesis contributes to a feature engineering phase, where myocardial segment features are extracted and used to train several traditional machine learning models and one-class classification methods reaching the highest sensitivity levels of 88.75% in single-view and 90.91% in multi-view echocardiography. The assessment of pulmonary regions plays a vital role in COVID-19 clinical diagnosis and prevention of contamination to the public. This thesis contributes to the composition of the pioneer and the largest CXR dataset, called QaTa-COV19, for the recognition, early detection, and pneumonia segmentation of COVID-19. Accordingly, another contribution of this thesis is the human-machine collaboration approach for the annotation of COVID-19 pneumonia regions for the segmentation. The thesis contributes to several machine learning models and solutions to perform a reliable detection and localization of COVID-19 pneumonia in the lungs using segmentation neural networks, direct detection from a proposed operational neural network, and representation-based classification models. The findings of this thesis are promising in computer-aided diagnosis of MI and COVID-19, where the proposed solutions reached high sensitivity levels of 90.9% and 98.6%, respectively preventing subjective and expert-dependent assessments and detection in 2D echocardiography and CXR imaging. The last contribution of this thesis is the developed visualization and assistive tools, where color-coded trackings on echocardiography recordings, myocardial infarction displacement curves, COVID-19 infection maps, and machine learning-based detection and annotation tools serve medical doctors in medical diagnostics addressing visual enhancements over 2D echocardiography and CXR imaging.",
            "source_url": "https:\/\/openalex.org\/W7133005963",
            "pdf_url": "",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7133005963",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:52",
            "updated_at": "2026-11-15 16:52:52"
        },
        {
            "id": 780,
            "title": "Personalized Support Vector Machines for Privacy-Preserving Federated Learning",
            "title_ar": "",
            "title_en": "Personalized Support Vector Machines for Privacy-Preserving Federated Learning",
            "author": "Aleksei Ponomarenko-Timofeev",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2025,
            "language": "en",
            "research_domain": "computer-vision",
            "research_subdomain": "federated-learning",
            "categories": "machine-learning; computer-vision; responsible-ai; ai-applications; predictive-analytics; federated-learning; fairness; ai-cybersecurity",
            "keywords": "",
            "abstract": "The pervasive digitalization of the environment enables monitoring of real-world systems by collecting data from multiple diverse sources. Such data may correspond to problems such as network intrusion detection or predicting the state of the wireless channel. While machine learning (ML)-based solutions are increasingly employed in such tasks, centralized learning approaches often fall short in terms of scalability and privacy. A more practical alternative is offered by federated learning (FL), which enables collaborative model training without disclosing raw data. However, FL brings a new set of challenges such as heterogeneous data distributions, diverse computational resources, and privacy requirements, which are rarely considered together in the algorithm design. This dissertation addresses the heterogeneity of the data structure and distribution, the heterogeneity of the computational resources, and the requirements for privacy preservation. Motivated by the fact that such factors may be present simultaneously, this dissertation focuses on developing lightweight privacy-preserving personalized FL algorithms. The work aims to bridge the gap in the FL field by formulating a set of algorithms for learning classification\/regression models (i), proposing policies for operating in highly heterogeneous systems and assessing their applicability in different scenarios (ii), assessing the applicability of conventional privacy preservation measures (iii), and proposing novel privacy-preserving mechanisms and evaluating their applicability (iv). Assessments are based on the results collected from extensive experiments across a broad range of scenarios and data types. The numerical results demonstrate the effectiveness of the proposed methods. In systems with heterogeneous computational resources and data, the proposed algorithms consistently maintain performance on par with or better than the state-of-the-art. The training delay was observed to be up to 20 % lower compared with the state-of-the-art personalized FL algorithms. Moreover, the proposed policies for operating in highly heterogeneous systems retain fairness and leverage heterogeneity to improve the quality of the learned models. The positive effect of these policies was observed through faster model convergence and stable fairness. It was also determined that conventional privacy preservation measures, such as differential privacy mechanisms, may be applied to the system where the proposed algorithms are used. The utility of the models learned by the algorithms was assessed for these privacy preservation measures. The proposed privacy preservation mechanisms, based on multiplicative noise, were found to provide better privacy\/utility trade-offs compared to the conventional Gaussian mechanism within a certain operational region. These regions of higher utility were explored for diverse datasets (e.g., image-based, time-series). Multiple approaches to improving the utility of the learned model while retaining the privacy guarantees were studied based on the properties of the algorithms. The advantages of these approaches are demonstrated through both qualitative insights and quantitative evidence. Overall, this thesis addresses a key gap in the FL literature by providing lightweight, privacy-preserving, and personalized algorithms designed for heterogeneous wireless systems.",
            "source_url": "https:\/\/openalex.org\/W7132950646",
            "pdf_url": "",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7132950646",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:53",
            "updated_at": "2026-11-15 16:52:53"
        },
        {
            "id": 893,
            "title": "Visually attended sound source semantic segmentation : Forming a semantic mask using visual and audio cues",
            "title_ar": "",
            "title_en": "Visually attended sound source semantic segmentation : Forming a semantic mask using visual and audio cues",
            "author": "Petteri Nuotiomaa",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2024,
            "language": "en",
            "research_domain": "nlp",
            "research_subdomain": "speech-recognition",
            "categories": "machine-learning; deep-learning; computer-vision; nlp; predictive-analytics; multimodal; image-segmentation; speech-recognition",
            "keywords": "",
            "abstract": "Sound Source Separation from single input audio mixture is a problem that has had many different proposed solutions. Many of the modern solutions involve using machine learning and especially deep learning to achieve good separation results. The goal of this thesis is to develop a solution to the sound source semantic segmentation problem, evaluate the performance of the proposed method, and compare it with similar solutions The thesis goes over the theory behind the problem and explores some of the previous solutions used for this problem. During the background research, it became clear that there aren’t many solutions trying to solve the exact problem of sound source semantic segmentation. This is why the thesis examines solutions for different problems, that have some relation to the sound source semantic segmentation. This thesis proposes a new solution for sound source semantic segmentation. The solution is based on some of the groundbreaking research done in the sound source separation field. The proposed deep learning system has two networks, one for audio and one for visual information. The audio network is a U-Net type network and the visual network is based on the ResNet model of networks. The addition of visual information to the final separation is done by weighing the audio instrument masks with the probabilities of predicted instruments. The network uses common hyperparameters that are used in different works in the field. The loss function is slightly different than any of the previous works as it uses probabilities instead of hard labels for ground truth values. The results from the proposed system indicate that the system can separate simple audio sources from an audio mixture. The solution didn’t achieve wanted results in all test cases and specifically for duet audios the method had problems separating the instruments. The system showed some promise in the multimodality approach, but the better results for the multimodality approach could be accounted for by the different initial randomization of the network. The proposed solution achieved results that are better than the baseline and the most simple approaches to the same problem but failed to meet the standard separating quality that can be expected from a modern deep learning approach.",
            "source_url": "https:\/\/openalex.org\/W7133075142",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/155461\/2\/NuotiomaaPetteri.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7133075142",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:54",
            "updated_at": "2026-11-15 16:52:54"
        },
        {
            "id": 902,
            "title": "Enhancing Domain-Specific Automated Audio Captioning: A Study on Adaptation Techniques and Transfer Learning",
            "title_ar": "",
            "title_en": "Enhancing Domain-Specific Automated Audio Captioning: A Study on Adaptation Techniques and Transfer Learning",
            "author": "Charith Raghavaraju",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2023,
            "language": "en",
            "research_domain": "deep-learning",
            "research_subdomain": "transformers",
            "categories": "machine-learning; deep-learning; nlp; transfer-learning; transformers; speech-recognition",
            "keywords": "",
            "abstract": "Automated audio captioning is a challenging cross-modal task that takes an audio sample as input to analyze it and generate its caption in natural language as output.The existing datasets for audio captioning such as AudioCaps and Clotho encompass a diverse range of domains, with current proposed systems primarily focusing on generic audio captioning.This thesis delves into the adaptation of generic audio captioning systems to domain-specific contexts, simultaneously aiming to enhance generic audio captioning performance.The adaptation of the generic models to specific domains has been explored using two different techniques: complete fine-tuning of neural model layers and layer-wise fine-tuning within transformers.The process involves initial training with a generic captioning setup, followed by adaptation using domain-specific training data.In generic captioning, the process for training starts with training the model on the Audio-Caps dataset followed by fine-tuning it using the Clotho dataset.This is accomplished through the utilization of a transformer-based architecture, which integrates a patchout fast spectrogram transformer (PaSST) for audio embeddings and a BART transformer.Word embeddings are generated using a byte-pair encoding (BPE) tokenizer tailored to the training datasets' unique words, aligning the vocabulary with the generic captioning task.Experimental adaptation mainly focuses on audio clips related to animals and vehicles.The results demonstrate notable improvements in the performance of the generic and domain adaptation systems.Generic captioning has demonstrated an improvement in SPIDEr scores, increasing from 0.291 during fine-tuning to 0.301 with layer-wise fine-tuning.Specifically, we observed a notable increase in SPIDEr scores, from 0.315 to 0.323 for animal-related audio clips and from 0.298 to 0.308 for vehicle-related audio clips.",
            "source_url": "https:\/\/openalex.org\/W7133068187",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/153190\/2\/RaghavarajuCharitha.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7133068187",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:54",
            "updated_at": "2026-11-15 16:52:54"
        },
        {
            "id": 926,
            "title": "Artificial Intelligence in Modern Firewalls : Role of Artificial Intelligence in Network Security",
            "title_ar": "",
            "title_en": "Artificial Intelligence in Modern Firewalls : Role of Artificial Intelligence in Network Security",
            "author": "Wajeh Ahmad",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2024,
            "language": "en",
            "research_domain": "deep-learning",
            "research_subdomain": "xai",
            "categories": "machine-learning; deep-learning; reinforcement-learning; responsible-ai; expert-systems; ai-applications; data-mining; xai; fairness; ai-cybersecurity; ai-smart-cities",
            "keywords": "",
            "abstract": "This thesis explores the role and impact of Artificial Intelligence on the design of modern firewalls due to the advancement of Artificial Intelligence (AI), particularly how AI-based methods transcend the barriers of the traditional rule-based approaches and enhance overall network security. The research traces the evolution of firewalls over time and highlights the necessity for more advanced security measures. It also outlines the progress of AI, setting the stage for a deep dive into the role of AI in security systems. The research critically assesses the limitation of conventional firewalls, such as their inability to detect zero-day attacks and their vulnerability to inspect encrypted traffic. It is investigated that how machine learning techniques, such as shallow neural networks or decision trees, can complement network security by aiding the firewall’s decision making and automatically classifying network packets. Furthermore, the thesis demonstrates the application of intrusion detection systems in cyberspace by employing machine learning, normal distribution, clustering, and t-plotting that can facilitate prevention and\/or detection of such malicious attacks. In addition, the thesis investigates the suitability of Explainable AI (XAI) to explain and make understandable AI based firewall decisions. It also highlights the difficulties and constraints of AI-centric firewalls, like the fact that they require a large set of data, the need of removing biases from such data, as well as taking privacy into account when using these capabilities, pushing further research on development of robust training methodologies and privacy-preserving approaches. The thesis features two prominent case studies: Google Chronicle Security Operations Platform, and Darktrace Enterprise Immune System. These case-studies highlight AI can be put to practical use for improving the abilities of threat detection, investigation and response. The thesis concludes by highlighting the significant progress made in integrating AI into contemporary firewalls, such that they can start learning proactively against continuous adapting cyber threat landscape. The report concludes by suggesting for more research to be done in the refining of AI algorithms, the investigation into reinforcement learning, and adding user and entity behavior analytics (UEBA) onto models as well as continuing the process of creating explainable AI models. The findings from this work will contribute to a greater understanding of AI-powered network security and offer direction for those looking at how they might develop more resilient firewalls in the future both in practice as well as in research.",
            "source_url": "https:\/\/openalex.org\/W7132940295",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/161576\/2\/AhmadWajeh.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7132940295",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:54",
            "updated_at": "2026-11-15 16:52:54"
        },
        {
            "id": 980,
            "title": "Large Language Models on Software Refactoring",
            "title_ar": "",
            "title_en": "Large Language Models on Software Refactoring",
            "author": "Mikko Metsola",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2024,
            "language": "en",
            "research_domain": "nlp",
            "research_subdomain": "llm",
            "categories": "nlp; llm",
            "keywords": "",
            "abstract": "The recent surge in popularity of large language models (LLM) such as OpenAI's GPT series of models has raised interest in the application of artificial intelligence (AI) in different fields, including software engineering. AI-powered programming assistants and code generation tools, namely GitHub Copilot, have decisively paved their way into common integrated development environments to provide feedback for software developers with programming tasks. These tasks may range from producing new software code, to providing suggestions for a programming task, to parsing through written code and analyzing it for potential improvements. Maintaining existing software is a very resource intensive task. Software programs are rarely written in a way that do not need adjustments later on. Addressing bugs and introducing new features are common tasks for developers, which require a reasonable understanding of the code they are working on. Refactoring is the process of rewriting software code in an attempt to streamline further development and maintenance or to improve the underlying design of the software without making changes to the overall functionality of the software. The thesis begins with an introduction to the topic of software quality. This aims to give a thorough understanding on which factors contribute to the quality of software. An overview of LLMs is presented as well as how they are generally applied to code generation. This thesis explores the extent to which a selection of four different LLMs can be utilized in code refactoring and what kinds of refactorings these LLMs are able to determine. This research is done by giving each LLM a piece of code to refactor. The outputs of these models are then evaluated on a variety of software metrics and commonly accepted code refactoring practices. With these results we can assess whether these language models could be applied in practice.",
            "source_url": "https:\/\/openalex.org\/W7132968924",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/160496\/2\/MetsolaMikko.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7132968924",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:56",
            "updated_at": "2026-11-15 16:52:56"
        },
        {
            "id": 1193,
            "title": "Application of Machine Learning Methods on Predictive Maintenance",
            "title_ar": "",
            "title_en": "Application of Machine Learning Methods on Predictive Maintenance",
            "author": "Feranmi Akanni",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2019,
            "language": "en",
            "research_domain": "machine-learning",
            "research_subdomain": "predictive-analytics",
            "categories": "machine-learning; ai-applications; predictive-analytics; ai-industry",
            "keywords": "",
            "abstract": "Maintenance is important in optimization of the business value of a functional unit and this optimization can only be achieved through predictive maintenance which ensures that maintenance activities are not carried out before due time and at the same time, prevent occurrence of breakdown of functional units because of missed maintenance activities.In this thesis, we focus on using different machine learning methods to predict the failure of a functional units.We explore the data and use missing data techniques to deal with missing values in the dataset, which resulted in a complete dataset.We explore various feature selection techniques to extract important features and reduce dimensionality of the dataset.Then, we explore the following machine learning methods: logistic regression, naïve Bayes, support vector machine, k-nearest neighbour and ensemble learning techniques which are bagging and boosting methods.Our results indicated that predictions from ensemble learning techniques have better evaluation metrics compared to other machine learning methods.",
            "source_url": "https:\/\/openalex.org\/W3010495304",
            "pdf_url": "",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W3010495304",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:52:59",
            "updated_at": "2026-11-15 16:52:59"
        },
        {
            "id": 1257,
            "title": "Generating Healthcare Reports Using Natural Language Processing : Fine-tuning Finnish Language Models",
            "title_ar": "",
            "title_en": "Generating Healthcare Reports Using Natural Language Processing : Fine-tuning Finnish Language Models",
            "author": "Tuuli Laitinen",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2025,
            "language": "en",
            "research_domain": "nlp",
            "research_subdomain": "llm",
            "categories": "machine-learning; deep-learning; nlp; medical-ai; transfer-learning; cnn; transformers; llm",
            "keywords": "",
            "abstract": "Generating reports in different healthcare departments is an important factor in keeping up with different phenomena occurring in the everyday work. Examples of these are the number of operations done, complications happened during them, and used instruments. Different systems the healthcare workers use might have their own reporting systems, which is the case for Tampere heart hospital’s cardiology and cardiothoracic surgery departments with their quality monitoring and reporting system called Kardio. The current problem with Kardio has been limited possibilities to generate reports and therefore a solution was to try and develop a system where users could generate reports by natural language prompts. The goal was to train Finnish language models so that they could be used to identify named entities for cardiology operations. The results obtained have been used to investigate whether such a feature could be used as a part in generating new reports. The implementation of the functionality began by considering the components required for its development. This led to the conclusion that a tool capable of identifying certain entities from natural language text was necessary. There are no language models based on medical text and data for the Finnish language readily available and training a whole new language model was not necessary. Because of this, the choice was to fine-tune existing Finnish language models with data related to the field of cardiology. The chosen three language models to be fine-tuned were two BERT based, FinBERT and Finnish RoBERTa, and one CNN-based language model by spaCy. The results showed that there are no significant differences in the fine-tuning of the language models, but the F1-scores indicate that FinBERT and Finnish RoBERTa perform slightly better than spaCy. The F1-scores, ranging around 0.95 to 0.98, were very good compared to the fine-tuning of English language models, but this was possibly due to the very limited and self-generated training data. Based on the results obtained, the fine-tuned language model could be used in recognising entities needed for cardiology operation report generation after further development. Additional work should be done to improve the training data so that the language model recognises all the entities needed for reports, while ensuring that the language model performs well with a wide range of inputs.",
            "source_url": "https:\/\/openalex.org\/W7133049232",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/224856\/2\/LaitinenTuuli.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7133049232",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:53:00",
            "updated_at": "2026-11-15 16:53:00"
        },
        {
            "id": 1258,
            "title": "Automating Programming Assignment Grading with a Fine-Tuned Large Language Model",
            "title_ar": "",
            "title_en": "Automating Programming Assignment Grading with a Fine-Tuned Large Language Model",
            "author": "Matti Syrjanen",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2025,
            "language": "en",
            "research_domain": "nlp",
            "research_subdomain": "llm",
            "categories": "machine-learning; nlp; ai-applications; transfer-learning; llm; ai-industry",
            "keywords": "",
            "abstract": "Grading programming assignments is a labor-intensive process that requires significant effort from university personnel and is often influenced by subjective biases. Automated grading systems, particularly those leveraging artificial intelligence, could offer a promising solution to these challenges. This thesis examines the potential of large language models to reduce the grading workload while also standardizing the process. The product of this study was a fine-tuned OpenAI GPT model trained on past programming assignment submission-feedback pairings with feedback templates and grading instructions from the Programming 2 course run by Tampere University. The research focused on measuring the feedback given by the fine-tuned model in terms of consistency and differences between the feedback given by the corresponding human graders. The base GPT model was fine-tuned with a small dataset and default settings. Evaluation results suggest that further experimentation with different hyperparameters and larger or alternative datasets is necessary to increase the reliability of the model. While the AI-generated feedback met predefined evaluation metrics in most cases, the results were not yet reliable enough for actual grading.",
            "source_url": "https:\/\/openalex.org\/W7132916378",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/225407\/2\/SyrjanenMatti.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7132916378",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:53:00",
            "updated_at": "2026-11-15 16:53:00"
        },
        {
            "id": 1260,
            "title": "An exploration of fine-tuning BERT for hate speech analysis",
            "title_ar": "",
            "title_en": "An exploration of fine-tuning BERT for hate speech analysis",
            "author": "Yingqi Zhao",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2024,
            "language": "en",
            "research_domain": "nlp",
            "research_subdomain": "llm",
            "categories": "machine-learning; deep-learning; nlp; transfer-learning; transformers; llm; sentiment-analysis; speech-recognition",
            "keywords": "",
            "abstract": "In the realm of Natural Language Processing (NLP), sentiment classification stands as an essential subject, commonly categorizing text into three classes: positive, negative, and neutral, enabling the statistical analysis of the emotional tendencies in the text.With the development of social networks, the classification requirements for sentiment analysis have gradually become more refined.This thesis draws datasets related to hate speech, fine-tuning BERT as a classification model to categorize the text into five categories to enhance the analysis of hate speech within social media platforms.Simultaneously, to address the substantial parameter size of large language models and the associated high fine-tuning costs, the thesis explores three methods for fine-tuning the BERT model: Fine-tuning BERT with all parameters, Fine-tuning BERT only the classification heads and Fine-tuning BERT with LoRA.This Thesis empirically validates the feasibility and superiority of designing targeted sentiment analysis models for refined scenarios.Achieving 90% accuracy confirms the outstanding performance of the BERT model in addressing hate speech analysis issues.Furthermore, comparing different fine-tuning methods validates the potential and value of the PEFT method for fine-tuning transformer models.",
            "source_url": "https:\/\/openalex.org\/W7133093837",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/155728\/7\/Zhao%20Yingqi.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7133093837",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:53:00",
            "updated_at": "2026-11-15 16:53:00"
        },
        {
            "id": 1261,
            "title": "Improving Code Quality Using Fine-tuning Large Language Models",
            "title_ar": "",
            "title_en": "Improving Code Quality Using Fine-tuning Large Language Models",
            "author": "Quang Hieu Nguyen",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2024,
            "language": "en",
            "research_domain": "nlp",
            "research_subdomain": "llm",
            "categories": "machine-learning; deep-learning; nlp; generative-ai; reinforcement-learning; transfer-learning; transformers; llm",
            "keywords": "",
            "abstract": "Large language models has demonstrated significant capabilities in solving real-life problems by the means of generating human-like responses to input text in the similar format. However, these generic languages exhibit weakness when the context is not fully provided, resulting a need for model customization to enhance their performance in specialized fields. The fine-tuning is implemented using techniques inspired by machine learning methods. The two well-known fine-tuning methods are Supervised Fine-Tuning (STF), based on the concept of supervised learning, and Reinforcement Learning from Human Feedback (RLHF) introduced by OpenAI. The resource-intensive nature of RLHF makes it challenging to replicate the results achieved by Chat Generative Pre-trained Transformer (ChatGPT) models. Therefore, a new technique named direct preference optimization (DPO) has been introduced lately, incorporating similar concepts to RLHF while reducing required time and labor. Given several studies have not addressed the improvement of DPO compared to other fine-tuning methods and its interplay with them, this thesis employs STF, DPO and their combination to fine-tune the base LLAMA 3.1 model, whose size is 8 billion parameters, in order to improve code quality generated by the LLMs. These fine-tuned models are then evaluated using Bi-Lingual Evaluation Understudy (BLEU) and Bidirectional Encoder Representations from Transformer (BERT) metrics, along with their code-specific derivatives. Despite the outcomes of the metrics cannot determine the most effective fine-tuning technique, they clearly highlight the significant influence of dataset utilized during the fine-tuning process. The repository containing the code and data for the research can be accessed via GitHub platform.",
            "source_url": "https:\/\/openalex.org\/W7133076851",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/161629\/2\/NguyenDuc.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7133076851",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:53:00",
            "updated_at": "2026-11-15 16:53:00"
        },
        {
            "id": 1285,
            "title": "Image Denoising via Deep Learning Neural Networks and Transformers",
            "title_ar": "",
            "title_en": "Image Denoising via Deep Learning Neural Networks and Transformers",
            "author": "Jundong Wang",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2024,
            "language": "en",
            "research_domain": "deep-learning",
            "research_subdomain": "transformers",
            "categories": "deep-learning; computer-vision; transformers",
            "keywords": "image denoising, deep learning, residual learning, Transformers.",
            "abstract": "This thesis investigates the implementation of image denoising using Neural Networks (NN) in conjunction with wavelet transform techniques. For the purpose of data simulation, images are intentionally corrupted with additive white Gaussian noise (AWGN) at various noise levels. Additionally, the study includes the examination of both random noise and real-world noise scenarios. This research has an emphasis on implementing several novel neural network architectures and strategies. By integrating the wavelet transform, the research aims to leverage its multi-resolution analysis capability to further enhance denoising performance. The proposed methods underscore the ability of residual learning and Transformers to integrate both local and global features, which efficiently capture fine-grained local details and long-range global dependencies. Keywords: image denoising, deep learning, residual learning, Transformers.",
            "source_url": "https:\/\/openalex.org\/W7133094216",
            "pdf_url": "https:\/\/trepo.tuni.fi\/bitstream\/10024\/160429\/2\/WangJundong.pdf",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W7133094216",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:53:00",
            "updated_at": "2026-11-15 16:53:00"
        },
        {
            "id": 1294,
            "title": "Customer Move Prediction In Electricity Sales Using Deep Learning Neural Networks",
            "title_ar": "",
            "title_en": "Customer Move Prediction In Electricity Sales Using Deep Learning Neural Networks",
            "author": "J. Laakso",
            "supervisor": "",
            "degree_type": "unspecified",
            "degree_evidence": "",
            "university": "Tampere University",
            "university_source": "repository_host",
            "college": "",
            "department": "",
            "country": "Finland",
            "governorate": "",
            "publication_year": 2020,
            "language": "en",
            "research_domain": "deep-learning",
            "research_subdomain": "predictive-analytics",
            "categories": "machine-learning; deep-learning; predictive-analytics",
            "keywords": "",
            "abstract": "",
            "source_url": "https:\/\/openalex.org\/W3035970992",
            "pdf_url": "",
            "doi": "",
            "handle": "",
            "repository_name": "Tampere University Institutional Repository (Tampere University)",
            "source_record_id": "W3035970992",
            "other_sources": "",
            "verification_status": "needs_review",
            "created_at": "2026-11-15 16:53:00",
            "updated_at": "2026-11-15 16:53:00"
        }
    ]
}