Lameski, Petre

Preferred name

Lameski, Petre

Official Name

Lameski, Petre

Main Affiliation

petre.lameski@finki.ukim.mk

Now showing 1 - 10 of 140

Some of the metrics are blocked by your
consent settings
Automating Feature Extraction from Entity-Relation Models: Experimental Evaluation of Machine Learning Methods for Relational Learning
(MDPI AG, 2024-04-01)
Stanoev, Boris
;
Mitrov, Goran
;
;
;
<jats:p>With the exponential growth of data, extracting actionable insights becomes resource-intensive. In many organizations, normalized relational databases store a significant portion of this data, where tables are interconnected through some relations. This paper explores relational learning, which involves joining and merging database tables, often normalized in the third normal form. The subsequent processing includes extracting features and utilizing them in machine learning (ML) models. In this paper, we experiment with the propositionalization algorithm (i.e., Wordification) for feature engineering. Next, we compare the algorithms PropDRM and PropStar, which are designed explicitly for multi-relational data mining, to traditional machine learning algorithms. Based on the performed experiments, we concluded that Gradient Boost, compared to PropDRM, achieves similar performance (F1 score, accuracy, and AUC) on multiple datasets. PropStar consistently underperformed on some datasets while being comparable to the other algorithms on others. In summary, the propositionalization algorithm for feature extraction makes it feasible to apply traditional ML algorithms for relational learning directly. In contrast, approaches tailored specifically for relational learning still face challenges in scalability, interpretability, and efficiency. These findings have a practical impact that can help speed up the adoption of machine learning in business contexts where data is stored in relational format without requiring domain-specific feature extraction.</jats:p>
Some of the metrics are blocked by your
consent settings
Deep Learning Methods for Bug Bite Classification: An End-to-End System
(MDPI AG, 2023-04-21)
;
Trojachanec Dineva, Katarina
;
Tojtovska Ribarski, Biljana
;
Petrov, Petar
;
Mladenovska, Teodora
<jats:p>A bite from a bug may expose the affected person to serious, life-threatening conditions, which may require immediate medical attention. The identification of the bug bite may be challenging even for experienced medical personnel due to the different manifestations of the bites and similarity to other skin conditions. This motivated our work on a computer-aided system that offers information on the bug bite based on the classification of bug bite images. Recently, there have been significant advances of methods for image classification for the detection of various skin conditions. However, there are very few sources that discuss the classification of bug bites. The goal of our research is to fill in this gap in the literature and offer a comprehensive approach for the analysis of this topic. This includes (1) the creation of a dataset that is larger than those considered in the related sources; (2) the exploration and analysis of the application of pre-trained state-of-the-art deep learning architectures with transfer learning, used in this study to overcome the challenges of low-size datasets and computational burden; (3) the further improvement of the classification performance of the individual CNNs by proposing an ensemble of models, and finally, (4) the implementation and description of an end-to-end system for bug bite classification from images taken with mobile phones, which should be beneficial to the medical personnel in the diagnostic process. In this paper, we give a detailed discussion of the models’ architecture, back-end architecture, and performance. According to the general evaluation metrics, DenseNet169 with an accuracy of 78% outperformed the other individual CNN models. However, the overall best performance (accuracy of 86%) was achieved by the proposed stacking ensemble model. These results are better than the results in the limited related work. Additionally, they show that deep CNNs and transfer learning can be successfully applied to the problem of the classification of bug bites.</jats:p>
Some of the metrics are blocked by your
consent settings
Temporal Authorization Graphs: Pros, Cons and Limits
(Springer International Publishing, 2022-01)
;
Popovski, Ognen
;
;
;
As more private data is entering the web, defining authorization about its access is crucial for privacy protection. This paper proposes a policy language that leverages SPARQL expressiveness and popularity for flexible access control management and enforces the protection using temporal graphs. The temporal graphs are created during the authentication phase and are cached for further usage. They enable design-time policy testing and debugging, which is necessary for correctness guarantee. The security never comes with convenience, and this paper examines the environments in which the temporal graphs are suitable. Based on the evaluation results, an approximated function is defined for suitability determination based on the expected temporal graph size.
Some of the metrics are blocked by your
consent settings
Enhancing Portfolio Management Using Artificial Intelligence: Literature Review
(Frontiers, 2024-03-12)
Sutiene, Kristina
;
Schwendner, Peter
;
Sipos, Ciprian
;
Lorenzo, Luis
;
Building an investment portfolio is a problem that numerous researchers have addressed for many years. The key goal has always been to balance risk and reward by optimally allocating assets such as stocks, bonds, and cash. In general, the portfolio management process is based on three steps: planning, execution, and feedback, each of which has its objectives and methods to be employed. Starting from Markowitz's mean-variance portfolio theory, different frameworks have been widely accepted, which considerably renewed how asset allocation is being solved. Recent advances in artificial intelligence provide methodological and technological capabilities to solve highly complex problems, and investment portfolio is no exception. For this reason, the paper reviews the current state-of-the-art approaches by answering the core question of how artificial intelligence is transforming portfolio management steps. Moreover, as the use of artificial intelligence in finance is challenged by transparency, fairness and explainability requirements, the case study of post-hoc explanations for asset allocation is demonstrated. Finally, we discuss recent regulatory developments in the European investment business and highlight specific aspects of this business where explainable artificial intelligence could advance transparency of the investment process.
Some of the metrics are blocked by your
consent settings
Digital Shift: Assessment of Mental States Through Passive Mobile Sensing
(Springer International Publishing, 2022)
Krajchevska, Evgenija
;
Petreska, Nina
;
Handjiski, Ognen
;
Andovska, Sandra
;
Some of the metrics are blocked by your
consent settings
Air pollution data: A dataset gathered through a crowd sensing platform
(2015-01-14)
Temkov, Slave
;
Chavkovski, Panche
;
;
;
This is a dataset on air pollution monitoring sourced from a crowd-sensing IoT platform. The dataset includes real-time data on various pollutants, including PM2.5, PM10, and NO2 levels, along with atmospheric data such as humidity and temperature. This data is collected across multiple urban locations in Skopje, North Macedonia.
Some of the metrics are blocked by your
consent settings
Technological Solutions for Older People with Alzheimer's Disease: Review
(Bentham Science Publishers Ltd., 2018-04-27)
Maresova, Petra
;
Tomsone, Signe
;
;
Madureira, Joana
;
Mendes, Ana
In the nineties, numerous studies began to highlight the problem of the increasing number of people with Alzheimer's disease in developed countries, especially in the context of demographic progress. At the same time, the 21st century is typical of the development of advanced technologies that penetrate all areas of human life. Digital devices, sensors, and intelligent applications are tools that can help seniors and allow better communication and control of their caregivers. The aim of the paper is to provide an up-to-date summary of the use of technological solutions for improving health and safety for people with Alzheimer's disease. Firstly, the problems and needs of senior citizens with Alzheimer's disease (AD) and their caregivers are specified. Secondly, a scoping review is performed regarding the technological solutions suggested to assist this specific group of patients. Works obtained from the following libraries used in this scoping review: Web of Science, PubMed, Springer, ACM and IEEE Xplore. Four independent reviewers screened the identified records and selected relevant articles which were published in the period from 2007 to 2018. A total of 6,705 publications were selected. In all, 128 full papers were screened. Results obtained from the relevant studies were furthermore divided into the following categories according to the type and use of technologies: devices, processing, and activity recognition. The leading technological solution in the category of devices are wearables and ambient non-invasive sensors. The introduction and utilization of these technologies however brings about challenges in acceptability, durability, ease of use, communication, and power requirements. Furthermore, in needs to be pointed out that these technological solutions should be based on open standards.
Some of the metrics are blocked by your
consent settings
Forecasting air pollution with deep learning with a focus on impact of urban traffic on PM10 and noise pollution
(Public Library of Science (PLoS), 2024)
Kostadinov, Martin
;
;
;
Coelho, Paulo Jorge
;
Air pollution constitutes a significant worldwide environmental challenge, presenting threats to both our well-being and the purity of our food supply. This study suggests employing Recurrent Neural Network (RNN) models featuring Long Short-Term Memory (LSTM) units for forecasting PM10 particle levels in multiple locations in Skopje simultaneously over a time span of 1, 6, 12, and 24 hours. Historical air quality measurement data were gathered from various local sensors positioned at different sites in Skopje, along with data on meteorological conditions from publicly available APIs. Various implementations and hyperparameters of several deep learning models were compared. Additionally, an analysis was conducted to assess the influence of urban traffic on air and noise pollution, leveraging the COVID-19 lockdown periods when traffic was virtually non-existent. The outcomes suggest that the proposed models can effectively predict air pollution. From the urban traffic perspective, the findings indicate that car traffic is not the major contributing factor to air pollution.
Some of the metrics are blocked by your
consent settings
Methods for urban Air Pollution measurement and forecasting: Challenges, opportunities, and solutions
(MDPI, 2023-09-15)
Mitreska Jovanovska, Elena
;
Batz, Victoria
;
;
;
Herzog, Michael A
In today’s urban environments, accurately measuring and forecasting air pollution is crucial for combating the effects of pollution. Machine learning (ML) is now a go-to method for making detailed predictions about air pollution levels in cities. In this study, we dive into how air pollution in urban settings is measured and predicted. Using the PRISMA methodology, we chose relevant studies from well-known databases such as PubMed, Springer, IEEE, MDPI, and Elsevier. We then looked closely at these papers to see how they use ML algorithms, models, and statistical approaches to measure and predict common urban air pollutants. After a detailed review, we narrowed our selection to 30 papers that fit our research goals best. We share our findings through a thorough comparison of these papers, shedding light on the most frequently predicted air pollutants, the ML models chosen for these predictions, and which ones work best for determining city air quality. We also take a look at Skopje, North Macedonia’s capital, as an example of a city still working on its air pollution measuring and prediction systems. In conclusion, there are solid methods out there for air pollution measurement and prediction. Technological hurdles are no longer a major obstacle, meaning decision-makers have ready-to-use solutions to help tackle the issue of air pollution.
Some of the metrics are blocked by your
consent settings
Automating feature extraction from entity-relation models: Experimental evaluation of machine learning methods for relational learning
(MDPI, 2024-04-01)
Stanoev, Boris
;
Mitrov, Goran
;
;
Mirceva, Georgina
;
With the exponential growth of data, extracting actionable insights becomes resource-intensive. In many organizations, normalized relational databases store a significant portion of this data, where tables are interconnected through some relations. This paper explores relational learning, which involves joining and merging database tables, often normalized in the third normal form. The subsequent processing includes extracting features and utilizing them in machine learning (ML) models. In this paper, we experiment with the propositionalization algorithm (i.e., Wordification) for feature engineering. Next, we compare the algorithms PropDRM and PropStar, which are designed explicitly for multi-relational data mining, to traditional machine learning algorithms. Based on the performed experiments, we concluded that Gradient Boost, compared to PropDRM, achieves similar performance (F1 score, accuracy, and AUC) on multiple datasets. PropStar consistently underperformed on some datasets while being comparable to the other algorithms on others. In summary, the propositionalization algorithm for feature extraction makes it feasible to apply traditional ML algorithms for relational learning directly. In contrast, approaches tailored specifically for relational learning still face challenges in scalability, interpretability, and efficiency. These findings have a practical impact that can help speed up the adoption of machine learning in business contexts where data is stored in relational format without requiring domain-specific feature extraction.

Lameski, Petre

Filters

Settings

Sort By

Results per page