Gievska, Sonja

Preferred name

Gievska, Sonja

Official Name

Gievska, Sonja

Main Affiliation

sonja.gievska@finki.ukim.mk

Now showing 1 - 10 of 38

Some of the metrics are blocked by your
consent settings
A Study of Different Models for Subreddit Recommendation Based on User-Community Interaction
(Springer, 2019-10)
Janchevski, Andrej
;
Abstract. Reddit is a community-oriented social network, where users can pose questions, share their own views and experiences within subred- dit communities they have subscribed to, with the possibility that other users might view, rate and comment on their posts. A recommender sys- tem plays a crucial role in advancing and steering interactions on social media platforms, and in the case of Reddit, it performs across many levels. This study investigates the potential benefits of social media analytics for improving the quality of recommendations. Five models are proposed and validated, with a particular focus on improving the recommendations of subreddits that might be of interest to a particular user. The results reinforce the notion that capturing and fusing diverse set of features is crucial for confronting the challenges of predicting elusive phenomenon such as user’s preferences and interests.
Some of the metrics are blocked by your
consent settings
Boosting Recommender Systems with Advanced Embedding Models
(ACM, 2020-04-19)
Cenikj, Gjorgjina
;
Recommender systems are paramount in providing personalized content and intelligent content filtering on any social media platform, web portal, and online application. In line with the current trends in the field directed towards mapping problem and data encoding representations from other fields, this research investigates the feasibility of augmenting a graph-based recommender system for Amazon products with two state-of-the-art representation models. In particular, the potential benefits of using the language representation model BERT and GraphSage based representations of nodes and edges for improving the quality of the recommendations were investigated. Link prediction and link attribute inference were used to identify the products that the users will buy and predict the rating they will give to a product, respectively. The initial results of our exploratory study are encouraging and point to potential directions for future research.
Some of the metrics are blocked by your
consent settings
Fake News Detection by Using Doc2Vec Representation Model and Various Classification Algorithms
(IEEE, 2021)
Janakieva, Dobrinka
;
;
Some of the metrics are blocked by your
consent settings
A Survey of Graph Neural Network Architectures in Ligand Binding Affinity Prediction Models
(IEEE, 2024-05-20)
Fetaji, Fjolla
;
;
Ligand affinity prediction plays a pivotal role in drug discovery, influencing the efficiency and success of drug development processes. Traditional methods struggle in accurately capturing the complex interactions within molecular structures, prompting the exploration of advanced techniques such as Graph Neural Networks (GNNs). This paper provides an analysis of GNNs in the context of ligand affinity prediction, exploring their architecture, applications, and potential impact on revolutionizing drug discovery. Our findings suggest that GNNs can offer improvements over traditional computational methods, particularly in handling the dynamic and complex nature of molecular interactions. We highlight innovative GNN architectures that have shown notable success in predicting ligand binding affinities, such as heterogeneous graph representation and attention mechanisms. The implications of these advancements suggest a paradigm shift in drug discovery, where GNNs can lead to more accurate predictions and accelerate the identification of potential drug candidates. This study underscores the transformative potential of GNNs in enhancing predictive accuracy and efficiency in drug development.
Some of the metrics are blocked by your
consent settings
A Complete Air Pollution Monitoring and Prediction Framework
(Institute of Electrical and Electronics Engineers (IEEE), 2023)
Kalajdjieski, Jovan
;
;
Mirceva, Georgina
;
;
Some of the metrics are blocked by your
consent settings
An Exploration of Autism Spectrum Disorder Classification from Structural and Functional MRI Images
(Springer Nature Switzerland, 2022-09-29)
Krajevski, Jovan
;
;
;
;
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 your
consent settings
Hyperparameter Optimization of Graph Neural Networks for mRNA Degradation Prediction
(IEEE, 2023-05-22)
Vodilovska, Viktorija
;
;
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 your
consent settings
Real-time Macedonian Sign Language Recognition System by using Transfer Learning
(IEEE, 2022-05-23)
Kralevska, A
;
Trajanov, R
;
The objective for developing a real-time sign language recognition system is twofold: improving inter-personal communication and supporting inclusive human-computer interaction with hearing-impaired population using a particular sign language. This study describes the design and implementation of a system for real-time Macedonian Sign Language recognition in images and videos. A robust and lightweight model was proposed based on transfer learning of suitable pretrained architectures, namely, Single Shot Detector (SSD) MobileNetV2 and SSD MobileNetV2 FPNLite. The proposed models were fine-tuned and extensively evaluated in a number of diverse scenarios to account for the inherent difficulties in recognizing particular letters. In the absence of publicly available dataset, we have created a dataset consisting of two-handed images of 28 out of 31 letters of the Macedonian alphabet; the three letters expressed by dynamic gestures were excluded from the study. The results point out to a state-of-the-art prediction accuracy on the classification task of Macedonian sign language alphabet.
Some of the metrics are blocked by your
consent settings
Explorations into Deep Learning Text Architectures for Dense Image Captioning
(2020-09)
;
Stojanovska, Frosina
;
;
;
Some of the metrics are blocked by your
consent settings
Time series anomaly detection with Variational Autoencoder using Mahalanobis distance
(Springer, 2020-09)
Gjorgjiev, L.
;
Two themes have dominated the research on anomaly de- tection in time series data, one related to explorations of deep architec- tures for the task, and the other, equally important, the creation of large benchmark datasets. In line with the current trends, we have proposed several deep learning architectures based on Variational Autoencoders that have been evaluated for detecting cyber-attacks on water distribution system on the BATADAL challenge task and dataset. The second research aim of this study was to examine the impact of using Mahalanobis distance as a reconstruction error on the performance of the proposed models.

Gievska, Sonja

Filters

Settings

Sort By

Results per page