Faculty of Electrical Engineering and Information Technologies

Permanent URI for this communityhttps://repository.ukim.mk/handle/20.500.12188/10

Browse

Search Results

Now showing 1 - 6 of 6

Some of the metrics are blocked by your
consent settings
A Survey of Bias in Healthcare: Pitfalls of Using Biased Datasets and Applications
(Springer, Cham, 2023-07-09)
Velichkovska, Bojana
;
;
Gjoreski, Hristijan
;
;
Osmani, Venet
Artificial intelligence (AI) is widely used in medical applications to support outcome prediction and treatment optimisation based on collected patient data. With the increasing use of AI in medical applications, there is a need to identify and address potential sources of bias that may lead to unfair decisions. There have been many reported cases of bias in healthcare professionals, medical equipment, medical datasets, and actively used medical applications. These cases have severely impacted the quality of patients’ healthcare, and despite awareness campaigns, bias has persisted or in certain cases even exacerbated. In this paper, we survey reported cases of different forms of bias in medical practice, medical technology, medical datasets, and medical applications, and analyse the impact these reports have in the access and quality of care provided for certain patient groups. In the end, we discuss possible pitfalls of using biased datasets and applications, and thus, provide the reasoning behind the need for robust and equitable medical technologies.
Some of the metrics are blocked by your
consent settings
Network Traffic Analysis and Control by Application of Machine Learning
(Springer, Cham, 2023-07-09)
Bogoevski, Zlate
;
Jovanovski, Ivan
;
Velichkovska, Bojana
;
Efnusheva, Danijela
The purpose of this paper is to analyses how networks work, how data is transmitted, what information we get from each router during data transmission, getting to know the basics of machine learning and how to create models that will learn how networks work. By applying machine learning methods, results are obtained that show us the shape of a network. With different methods we can get information about how we can plan the network, in terms of expanding the network if the capacity of the links is almost full or when one of the links has predispositions to go from an active state to an inactive one. The results show satisfactory outcomes through the use of three different machine learning models that were capable of accurately detecting the functionality of a port, calculating its utilization and learning when the utilization hits a threshold of above 75%.
Some of the metrics are blocked by your
consent settings
Analysis of Early Cancer Diagnosis Using Machine Learning
(Springer, Cham, 2024)
Gjosheva, Marija
;
Bogoevski, Zlate
;
Velichkovska, Bojana
;
Efnusheva, Danijela
;
Cancer is a group of diseases with similar symptoms, all involving uncontrolled growth and reproduction of cells. With around 8 million deaths each year, it is the second leading cause of death worldwide in developing countries and the first in the developed world. In contemporary medicine, early cancer diagnosis for every known type is essential. Machine learning has the potential to completely transform the process and increase the number of lives saved. In order to make predictions, computers develop complex data models and search for patterns. Early cancer diagnosis could undergo a revolution because of machine learning. This research’s goal is to outline the issue surrounding cancer diagnoses in patients and all the difficulties they experience. A suitable strategy will be to model the risk of cancer and patient outcomes given the growing trend of employing machine learning technics in cancer research. A specific model has been developed that, if applied appropriately, can reduce the number of lost lives and, at the same time, increase the number of individuals capable of coping with this disease. The results indicate that the created model can be used by professionals to identify lung cancer with efficiency. If the prediction is accurate, the doctor may be able to develop a better treatment plan and provide the patient with an early diagnosis. The study's findings show that the number of patients has been rising recently, yet early detection is crucial because it can help avert serious complications.
Some of the metrics are blocked by your
consent settings
Mobile Edge Computing services with QoS support model for Next Generation Mobile Networks
(FEEIT, Skopje, 2022)
Shuminoski, Tomislav
;
Velichkovska, Bojana
;
This paper presents a novel overview in intelligent multi-access QoS mobile edge computing (MEC) for beyond 5G networks and services. There are many challenges faced by the expansion of Cloud networks and Mobile Networks, which can be solved by providing connectivity at the edge of the network, i.e. with Mobile Edge computing networks. The MEC improves overall network performance and reduces end-to-end service delay. Also, the improved advanced QoS model including Machine Learning (ML) algorithm within for next generation of mobile networks and services are proposed. The purpose of the ML algorithm is to understand the traffic activity and determine how the traffic schedule should be made. Given a set of machines and a set of jobs, the model should compute the processing schedule that minimizes specified metrics. The proposed model combines the most powerful features of both Cloud and Edge computing, independent from any existing and future Radio Access Technology, leading to possible better performance utility networks, lower service delay with high QoS provisioning for many used multimedia service. Finally, this paper gives an overview of the existing Mobile Edge Computing technologies and several existing use cases. Undoubtedly, MEC with QoS support is an innovative network paradigm going in 6G, which can essentially answer many of the existing Mobile Networks’ challenges.
Some of the metrics are blocked by your
consent settings
Machine Learning Based Classification of IoT Traffic
(Brno University of Technology, 2023-06)
Velichkovska, Bojana
;
;
With the rapid expansion and widespread adoption of the Internet of Things (IoT), maintaining secure connections among active devices can be challenging. Since IoT devices are limited in power and storage, they cannot perform complex tasks, which makes them vulnerable to different types of attacks. Given the volume of data generated daily, detecting anomalous behavior can be demanding. However, machine learning (ML) algorithms have proven successful in extracting complex patterns from big data, which has led to active applications in IoT. In this paper, we perform a comprehensive analysis, including 4 ML algorithms and 3 neural networks (NNs), and propose a pipeline which analyzes the influence data reduction (loss) has on the performance of these algorithms. We use random undersampling as a data reduction technique, which simulates reduced network traffic data. The pipeline investigates several degrees of data loss. The results show that models trained on the original data distribution obtain accuracy that verges on 100%. XGBoost performs best from the classic ML algorithms. From the deep learning models, the 2-layered NN provides excellent results and has sufficient depth for practical application. On the other hand, when the models are trained on the undersampled data, there is a decrease in performance, most notably in the case of NNs. The most prominent change is seen in the 4-layered NN, where the model trained on the original dataset detects attacks with a success of 93.53%, whereas the model trained on the maximally reduced data has a success of only 39.39%.
Some of the metrics are blocked by your
consent settings
Wearable Sensors Data-Fusion and Machine-Learning Method for Fall Detection and Activity Recognition
(Springer International Publishing, 2020)
Gjoreski, Hristijan
;
Stankoski, Simon
;
Kiprijanovska, Ivana
;
Nikolovska, Anastasija
;
Mladenovska, Natasha

Faculty of Electrical Engineering and Information Technologies

Browse

Filters

Settings

Sort By

Results per page

Search Results