Ivanoska, Ilinka
Preferred name
Ivanoska, Ilinka
Official Name
Ivanoska, Ilinka
Main Affiliation
Email
ilinka.ivanoska@finki.ukim.mk
24 results
Now showing 1 - 10 of 24
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Feature Selection for Improved Classification of Protein Structures(IEEE, 2019-05); ; - Some of the metrics are blocked by yourconsent settings
Item type:Publication, An Exploration of Autism Spectrum Disorder Classification from Structural and Functional MRI Images(Springer Nature Switzerland, 2022-09-29); ; ; There are strong indications that structural and functional magnetic resonance imaging (MRI) may help identify biologically relevant phenotypes of neurodevelopmental disorders such as Autism spectrum disorder (ASD). Extracting patterns from MRI data is challenging due to the high dimensionality, limited cardinality and data heterogeneity. In this paper, we explore structural and resting state functional MRI (rs-fMRI) for ASD classification using available ABIDE II dataset, using several standard machine learning (ML) models and convolutional neural networks (CNNs). To overcome the high dimensionality problem, we propose a simple data transformation method based on histograms calculation for the standard ML models and a simple 3D-to-2D and 4D-to-3D data transformation method for the CNNs in ASD classification. Numerous research has been done for ASD classification using the online available ABIDE I dataset, and several with the ABIDE II dataset, the latter mostly working with single-site classification studies. Here, we take the whole ABIDE II dataset using all structural and functional raw data. Our results show that the proposed methods achive state-of-the art results of high 71.4% accuracy (functional) and 73.4% AUC (structural) compared to the best performing results in literature of 68% accuracy (functional) for ASD classification on all ABIDE dataset. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Analysing the Influence of Two Similarity Metrics on the Ant Colony Optimisation Based Fuzzy-Rough Feature Selection Algorithm(IEEE, 2019-05); ; Mirceva, Georgina - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Hyperparameter Optimization of Graph Neural Networks for mRNA Degradation Prediction(IEEE, 2023-05-22); Graph Neural Networks (GNN) emerged as increasingly attractive deep learning models for complex data, making them extremely useful in biochemical and pharmaceutical domains. However, building a good-performing GNN requires lots of parameter choices and Hyperparameter optimization (HPO) can aid in exploring solutions. This study presents a comparative analysis of several strategies for Hyperparameter optimization of GNNs. The explored optimization techniques include complex algorithms such as the bio-inspired Genetic Algorithm, Particle Swarm Optimization, and Artificial Bee Colony. In addition, Hill Climb and Simulated Annealing as well as the commonly used methods Random Search and Bayesian Search have also been covered. The proposed optimization algorithms have been evaluated on improving the performance of the GNN architectures developed for predicting mRNA degradation. The Stanford OpenVaccine dataset for mRNA degradation prediction has been used for training and testing the predictive models. Finding mRNA molecules with low degradation rates is important in development of mRNA vaccines for diseases such as COVID-19 and we hope to benefit research on ML in this domain. According to the analysis’s findings, Simulated Annealing algorithm outperforms other algorithms on both architectures. Furthermore, population based algorithms like Particle Swarm optimization show promising results, with certain limitations related to the complexity of the algorithms which encourages further exploration of the subject. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Analysing the Influence of Two Similarity Metrics on the Ant Colony Optimisation Based Fuzzy-Rough Feature Selection Algorithm(IEEE, 2019-05); ; Mirceva, GeorginaAs the amount of the data increases in volume, veracity, and value, and thus the number of the features rises, so the standard algorithms will find it difficult to process the data without help of huge computer power. However, the feature selection methodology offers help for this issue. In its core, the feature selection tries to find the most predictive input features for the output (target) feature. Feature selection combined with a hybrid variant of rough sets, fuzzy-rough sets provides fuzzy-rough feature selection that could offer better results in this task. To help fuzzy-rough methods to find optimal subsets, attempts have been made using feature selection mechanism based on Ant Colony optimisation (ACO). This approach is applied to the problem of finding optimal subset of features in fuzzy-rough data reduction process by using different similarity metrics. In this paper, we investigate the influence of two fuzzy similarity metrics on the performance of the feature selection ACO strategy. The investigation is made by using several datasets. We experimentally compare several classical classification algorithms by using the AUC-ROC evaluation measure. Additionally, we show the benefits of making feature reduction. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Check for updates Multiplex Collaboration Network of the Faculty of Computer Science and Engineering in Skopje(Springer Nature, 2024); ; Multiplex collaboration networks facilitate intricate connections among individuals, enabling multidimensional collaborations across various domains and fostering synergistic knowledge exchange. This study focuses on the construction and basic analysis of a multiplex collaboration network among employees at the Faculty of Computer Science and Engineering (FCSE), Ss. Cyril and Methodius University in Skopje. The multiplex network is built with three layers based on: scientific collaborations resulting from joint project participations by FCSE employees, joint employees participations in the FCSE graduation thesis committees, and scientific FCSE employees collaborations defined by co-authorships in Google Scholar papers. The network's structure plays a vital role in determining the information accessibility and cooperative opportunities for individuals within FCSE institution. The aim here is to investigate the FCSE multiplex collaboration network's internal structure for discovering its latent knowledge and understand its implications. We perform identification of key individuals within the network, by computing various centrality and hubs detection network metrics. Additionally, we employ a community detection algorithm to reveal the underlying modular structure of the network. By comprehensively analyzing the acquired multiplex collaboration network model, we contribute to a better understanding of the collaboration patterns among FCSE employees. The findings can potentially inform decision-making processes and foster strategic planning aimed at enhancing collaboration and knowledge sharing within the institution. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Statistical and Machine Learning Link Selection Methods for Brain Functional Networks: Review and Comparison(Multidisciplinary Digital Publishing Institute, 2021-06); ; ; Zanin, MassimilianoNetwork-based representations have introduced a revolution in neuroscience, expanding the understanding of the brain from the activity of individual regions to the interactions between them. This augmented network view comes at the cost of high dimensionality, which hinders both our capacity of deciphering the main mechanisms behind pathologies, and the significance of any statistical and/or machine learning task used in processing this data. A link selection method, allowing to remove irrelevant connections in a given scenario, is an obvious solution that provides improved utilization of these network representations. In this contribution we review a large set of statistical and machine learning link selection methods and evaluate them on real brain functional networks. Results indicate that most methods perform in a qualitatively similar way, with NBS (Network Based Statistics) winning in terms of quantity of retained information, AnovaNet in terms of stability and ExT (Extra Trees) in terms of lower computational cost. While machine learning methods are conceptually more complex than statistical ones, they do not yield a clear advantage. At the same time, the high heterogeneity in the set of links retained by each method suggests that they are offering complementary views to the data. The implications of these results in neuroscience tasks are finally discussed. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Protein Function Prediction Using Semantic Similarity Metrics and Random Walk Algorithm(Faculty of Computer Science and Engineering, Ss. Cyril and Methodius University in Skopje, Macedonia, 2012); ; Most protein function prediction methods that have been proposed, are based on sequence or structure protein similarity and do not take into consideration the semantic similarity extracted from protein knowledge databases such as Gene Ontology. In this paper we present an approach for protein function prediction using semantic similarity metrics and the whole network topology of a protein interaction network by using a—semantic driven “random walk with restart. Different semantic similarity metrics are explored and future results should show the relevance of different semantic similarity metrics on protein function prediction using random walk with restart. To achieve the final goal of protein function prediction, the best semantic similarity metric should be used. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Assessing Identifiability in Airport Delay Propagation Roles Through Deep Learning Classification(IEEE, 2022-03-10); - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Automated Structural Classification of Proteins by Using Decision Trees and Structural Protein Features(Springer, Berlin, Heidelberg, 2009-09-28); ; ; ; The protein function is tightly related to classification of proteins in hierarchical levels where proteins share same or similar functions. One of the most relevant protein classification schemes is the structural classification of proteins (SCOP). The SCOP scheme has one negative drawback; due to its manual classification methods, the dynamic of classification of new proteins is much slower than the dynamic of discovering novel protein structures in the protein data bank (PDB). In this work, we propose two approaches for automated protein classification. We extract protein descriptors from the structural coordinates stored in the PDB files. Then we apply C4.5 algorithm to select the most appropriate descriptor features for protein classification based on the SCOP hierarchy. We propose novel classification approach by introducing a bottom-up classification flow, and a multi-level classification approach. The results show that these approaches are much faster than other similar algorithms with comparable accuracy.
- «
- 1 (current)
- 2
- 3
- »