Iskanje po katalogu digitalne knjižnice

Opis: Machine learning (ML) has demonstrated remarkable ability to uncover hidden patterns in data. However, the presence of biases and discrimination originating from the data itself and, consequently, emerging in the ML outcomes, remains a pressing concern. With the exponential growth of unstructured data, such as images, fairness has become increasingly critical, as neural network (NN) models may inadvertently learn and perpetuate societal and historical biases. To address this challenge, we propose a fairness-aware loss function that iteratively prioritizes the worst-performing sensitive group during NN training. This approach aims to balance treatment quality across sensitive groups, achieving fairer image classification outcomes while incurring only a slight compromise in overall performance. Our method, evaluated on the FairFace dataset, demonstrates significant improvements in fairness metrics while maintaining comparable overall quality. These trade-offs highlight that the minor decrease in overall quality is justified by the improvement in fairness of the models.
Ključne besede: fairness, search-basimage classification, machine learning, supervised learnign, neural networks
Objavljeno v DKUM: 23.04.2025; Ogledov: 0; Prenosov: 0
Celotno besedilo (2,01 MB)

Opis: Background and purpose: The aim of this study was to examine the risk factors, prodromal symptoms, non-motor symptoms (NMS), and motor symptoms (MS) in different Parkinson’s disease (PD) non-motor subtypes, classified using newly established criteria and a data-driven approach. Methods: A total of 168 patients with idiopathic PD underwent comprehensive NMS and MS examinations. NMS were assessed by the Non-Motor Symptom Scale (NMSS), Montreal Cognitive Assessment (MoCA), Hamilton Depression Scale (HAM-D), Hamilton Anxiety Rating Scale (HAM-A), REM Sleep Behavior Disorder Screening Questionnaire (RBDSQ), Epworth Sleepiness Scale (ESS), Starkstein Apathy Scale (SAS) and Fatigue Severity Scale (FSS). Motor subtypes were classified based on Stebbins’ method. Patients were classified into groups of three NMS subtypes (cortical, limbic, and brainstem) based on the newly designed inclusion criteria. Further, data-driven clustering was performed as an alternative, statistical learning-based classification approach. The two classification approaches were compared for consistency. Results: We identified 38 (22.6%) patients with the cortical subtype, 48 (28.6%) with the limbic, and 82 (48.8%) patients with the brainstem NMS PD subtype. Using a data-driven approach, we identified five different clusters. Three corresponded to the cortical, limbic, and brainstem subtypes, while the two additional clusters may have represented patients with early and advanced PD. Pearson chi-square test of independence revealed that a priori classification and cluster membership were significantly related to one another with a large effect size (χ2(8) = 175.001, p < 0.001, Cramer’s V = 0.722). The demographic and clinical profiles differed between NMS subtypes and clusters. Conclusion: Using the actuarial and clustering approach, marked differences between individual NMS subtypes were found. The newly established criteria have potential as a simplified tool for future clinical research of NMS subtypes of Parkinson’s disease.
Ključne besede: Parkinson’s disease, non-motor symptoms subtypes, a priori classification, cluster analysis
Objavljeno v DKUM: 07.04.2025; Ogledov: 0; Prenosov: 2
Celotno besedilo (1,37 MB)
Gradivo ima več datotek! Več...

Ključne besede: convolutional neural networks, generative adversarial networks, classification, synthetic images, explanable artificial intelligence
Objavljeno v DKUM: 29.01.2025; Ogledov: 0; Prenosov: 7
Celotno besedilo (36,84 MB)

Opis: The master’s degree thesis is composed of theoretical and practical parts. The theoretical part describes the basics of predictive data analytics and machine learning algorithms for classification such as Logistic Regression, Decision Tree, Random Forest, SVM, and KNN. We also describe different evaluation metrics such as Recall, Precision, Accuracy, F1 Score, Cohen’s Kappa, Hamming Loss, and Jaccard Index that are used to measure the performance of these algorithms. Additionally, we record the time taken for the training and prediction processes to provide insights into algorithm scalability. The key part master’s thesis is the practical part that compares these algorithms with a self-implemented tool that shows results for different evaluation metrics on seven datasets. First, we describe the implementation of an application for testing where we measure evaluation metrics scores. We tested these algorithms on all seven datasets using Python libraries such as scikit-learn. Finally, w
Ključne besede: data analytics, machine learning, classification, evaluation metrics
Objavljeno v DKUM: 15.01.2025; Ogledov: 0; Prenosov: 52
Celotno besedilo (2,68 MB)

Opis: Introduction: Type 2 diabetes mellitus is a major global health concern, but interpreting machine learning models for diagnosis remains challenging. This study investigates combining association rule mining with advanced natural language processing to improve both diagnostic accuracy and interpretability. This novel approach has not been explored before in using pretrained transformers for diabetes classification on tabular data. Methods: The study used the Pima Indians Diabetes dataset to investigate Type 2 diabetes mellitus. Python and Jupyter Notebook were employed for analysis, with the NiaARM framework for association rule mining. LightGBM and the dalex package were used for performance comparison and feature importance analysis, respectively. SHAP was used for local interpretability. OpenAI GPT version 3.5 was utilized for outcome prediction and interpretation. The source code is available on GitHub. Results: NiaARM generated 350 rules to predict diabetes. LightGBM performed better than the GPT-based model. A comparison of GPT and NiaARM rules showed disparities, prompting a similarity score analysis. LightGBM’s decision making leaned heavily on glucose, age, and BMI, as highlighted in feature importance rankings. Beeswarm plots demonstrated how feature values correlate with their influence on diagnosis outcomes. Discussion: Combining association rule mining with GPT for Type 2 diabetes mellitus classification yields limited effectiveness. Enhancements like preprocessing and hyperparameter tuning are required. Interpretation challenges and GPT’s dependency on provided rules indicate the necessity for prompt engineering and similarity score methods. Variations in feature importance rankings underscore the complexity of T2DM. Concerns regarding GPT’s reliability emphasize the importance of iterative approaches for improving prediction accuracy.
Ključne besede: GPT, association rule mining, classification, interpretability, diagnostics
Objavljeno v DKUM: 26.11.2024; Ogledov: 0; Prenosov: 223
Celotno besedilo (1,29 MB)
Gradivo ima več datotek! Več...