Ribarski, Panche
Preferred name
Ribarski, Panche
Official Name
Ribarski, Panche
Main Affiliation
Email
pance.ribarski@finki.ukim.mk
10 results
Now showing 1 - 10 of 10
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Stability of Coupled Systems of Stochastic Cohen-Grossberg Neural Networks with Time Delays, Impulses and Markovian Switching(Springer International Publishing, 2021); - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Application of Hierarchical Bayesian Model in Ophtalmological Study(Springer, 2019); ; Ljubic, AntonelaThe problems with statistical results based on p-values, together with multiple comparisons have been criticized often in the literature. Many authors argue that this way of reporting scientific research creates unreliable results. This issue is especially important in the era of Big Data, when many tests are done on the same data sets, which are often openly available. A way to overcome these problems is offered by Bayesian analysis. In our previous research we have used traditional statistical approach to conduct multiple hypothesis tests on our data in ophtalmological study. The goal of this paper is to apply the hierarchical Poisson exponential model on the data and test the dependence of congenital heart disease and Brusfield spots. We give detailed description of the model, analyze the generated Markov chains and the posterior distributions for the simulated parameters and discuss the results from Bayesian perspective. The results are original and have not been published yet. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, General decay stability analysis of coupled systems of stochastic neural networks(2019); This paper covers the topic of the pth moment (p ≥ 2) stability of coupled systems of stochastic Cohen-Grossberg neural networks with time delays, impulses and Markovian switching. This model generalises many models in the literature and to the best of our knowledge has not been analyzed before. The methods are based on results from graph theory, Lyapunov operator, Dini derivative and some known inequality techniques. Additionally, we consider the stability with respect to a general decay function which includes exponential, but also more general lower rate decay functions as the polynomial and the logarithmic ones. This fact gives us the opportunity to study general decay stability, even when the exponential one cannot be discussed. The presented theoretical results are supported by a numerical example. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, 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 yourconsent settings
Item type:Publication, On the General Decay Stability of Coupled System of Stochastic Neural Networks with Impulses, Markovian Switching and Node and Interconnection Delays(Springer International Publishing, 2024); - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Comparing Databases for Inserting and Querying Jsons for Big Data(2019); ; This paper tackles the topic of performance in Big Data data ingestion, data querying and data analytics. We test the import of the Last.fm Million Song Dataset in Cassandra, Mongo, PostgreSQL, CockroachDB, Mariadb and Elasticsearch. We also test three types of queries over the JSON documents and present the test results for unique visibility in the direct comparison of the performance of the selected databases. We conclude that by using a combination of state of the art scalable database, for which we recommend Cassandra, and Elasticsearch for search and analytics, we can get a unique tool for efficiently storing and querying data which can schema easily along the way. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Introduction To Secure Electronic Voting Requirements(Institute of Informatics, Faculty of Natural Sciences and Mathematics, Ss. Cyril and Methodius University in Skopje, Macedonia, 2010); Antovski, LjupchoVoting takes great part in the human society. The act of choosing a representative in a vote-like process has always been a delicate and fragile. Electronic voting is being introduced not so long ago [12]. We can see the progress of evoting just by looking the goals that researchers set their selves [1, 3, 7, 21, 22]. Accomplishment of these goals went in two directions: using blind signatures [1, 2, 3, 4, 5, 12, 24], or using homomorphic crypto-system [6, 9, 10, 20, 26]. The notion of mixnets is also intertwining with e-voting systems to ensure an anonymous communication channel. This research will try to unite these concepts and to review the various techniques brought by researchers in the past 20 years or so. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Mixnets: Implementation and performance evaluation of decryption and re-encryption types(Fakultet elektrotehnike i računarstva Sveučilišta u Zagrebu, 2012-09-30); Antovski, LjupchoThe anonymous channels have been the essence of numerous protocols that include anonymous message passing between peers. The mixnet structure is one way to accomplish the anonymity . Since the publication of the Chaumian mixnet, there have been many practical implementations. There are two main approaches to implement mixnets: the decryption (Chaumian) and the re-encryption mixnets. In this paper we analyze four types of mixnets, of which one decryption and three re-encryption types. They were implemented in the Java programing language and evaluated on several criterias as: the number of messages, the total number of nodes, the number of threshold nodes, and the key length of underlying crypto system. In the results section we compare the results from the practical tests to answer the research question, which type of mixnets has better features. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Deploying Production-Grade Kubernetes Cluster(2019-05); Deploying a production-grade Kubernetes cluster is a challenging feat, mainly because of all the different services that need to be integrated with each other. We examined Kubespray, which is an open-source project whose purpose is to automate the deployment of stable clusters, as well as ease the future administration and lifecycle management along the way. Additional components can be installed during the deployment process that enrich and improve the cluster capabilities. We used Kubespray and together with other open-sources projects we created production-grade Kubernetes cluster ready for real deployment scenarios. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Bayesian vs. frequentist inference - an ophtalmic study on ocular perfusion pressure(2024); ; ;Ljubic, AntonelaDimitrova, GalinaThis study investigates the application of Bayesian statistical methods for comparing ocular perfusion pressure (OPP) between glaucoma and non-glaucoma populations, contrasting it with traditional frequentist approaches. Using OPP measurements from two patient groups, we employ partially informed Bayesian models to test the hypothesis of no difference in means between the groups. We calculate Bayes factor using Savage-Dickey density ratio and offer insights in the hypothesis beyond p-values. The results highlight the advantages of the Bayesian approach, including its flexibility in incorporating prior information and interpreting evidence. We discuss the limitations and potential biases introduced by the choice of priors. This paper contributes to the understanding of Bayesian inference in ophthalmic research and emphasizes its potential for hypothesis testing in clinical studies.
