Testing the Semi Markov Model Using Monte Carlo Simulation Method for Predicting the Network Traffic

Kordnoori, Shirin; Mostafaei, Hamidreza; Kordnoori, Shaghayegh; Ostadrahimi, Mohammadmohsen

doi:10.18187/pjsor.v16i4.3394

Cited by 3 publications

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…HydraGAN's ensures the privacy of sensitive data by generating synthetic data that obfuscates sensitive information while retaining the predictive concepts within the original data. Synthetic data generation techniques vary widely in form, including probabilistic approaches to simulating epidemiological information [15,20,33], deep methods such as autoencoders that create synthetic medical images [40,44,46], and Markov models being that synthesize and predict network traffic [25].…”

Section: Synthetic Data Generationmentioning

confidence: 99%

HydraGAN A Multi-head, Multi-objective Approach to Synthetic Data Generation

DeSmet¹,

Cook²

2021

Preprint

View full text Add to dashboard Cite

Synthetic data generation overcomes limitations of real-world machine learning. Traditional methods are valuable for augmenting costly datasets but only optimize one criterion: realism. In this paper, we tackle the problem of generating synthetic data that optimize multiple criteria. This goal is necessary when real data are replaced by synthetic for privacy preservation. We introduce HydraGAN, a new approach to synthetic data generation that introduces multiple generator and discriminator agents into the system. The multi-agent GAN optimizes the goal of privacypreservation as well as data realism. To facilitate multi-agent training, we adapt game-theoretic principles to offer equilibrium guarantees. We observe that HydraGAN outperforms baseline methods for three datasets for multiple criteria of maximizing data realism, maximizing model accuracy, and minimizing re-identification risk.

show abstract

Section: Synthetic Data Generationmentioning

confidence: 99%

HydraGAN A Multi-head, Multi-objective Approach to Synthetic Data Generation

DeSmet¹,

Cook²

2021

Preprint

View full text Add to dashboard Cite

show abstract

Markov modeling on dynamic state space for genetic disorders and infectious diseases with mutations: Probabilistic framework, parameter estimation, and applications

Baranon,

Weke,

Alladatin

et al. 2024

Computational and Mathematical Biophysics

View full text Add to dashboard Cite

The emergence and dynamic prevalence of genetic disorders and infectious diseases with mutations pose significant challenges for public health interventions. This study investigated the parameter estimation approach and the application of the dynamic state-space Markov modeling of these conditions. Using extensive simulations, the model demonstrated robust parameter estimation performance, with biases and mean-squared errors decreasing as sample size increased. Applying the model to COVID-19 data revealed distinct temporal patterns for each variant, highlighting their unique emergence, peak dominance, and decline or persistence trajectories. Despite the absence of clear trends in the data, the model exhibited a remarkable accuracy in predicting future prevalence trends for most variants, showcasing its potential for real-time monitoring and analysis. While some discrepancies were observed for specific variants, these findings suggest the model’s promise as a valuable tool for informing public health strategies. Further validation with larger datasets and exploration of incorporating additional factors hold the potential for enhancing the model’s generalizability and applicability to other evolving diseases.

show abstract

Частная Полумарковская Модель Как Инструмент Снижения Сложности Задачи Оценивания Устойчивости Функционирования Элементов Информационной Инфраструктуры, Подверженной Воздействию Угроз

Voevodin

2024

View full text Add to dashboard Cite

Decision-making on information infrastructure (II) security for its sustainable functioning in the face of threats requires a tool to assess the sustainability of its individual elements. The application of the semi-Markov model to assess the stability of the functioning of elements of II exposed to threats in a direct setting is associated with the increasing complexity of the description of the object of delineation (parametric space) in step progression from the number of the considered impacts, which reduces its practical significance. However, no studies have been found in the scientific literature to reduce the complexity of the semi-Markov model. The article presents an approach to reducing the complexity of modeling by adopting correct assumptions when forming the initial data. Given the conditions under which it is possible to take a series. It was a cost to limit the applicability of the model by significantly reducing the complexity of the modelling. The problem statement and the modified transition graph are given. The novelty of the problem statement is to take into account the limitations on the available resources for the restoration of functionality of the element. To explain the physical essence of the modeling process, a thought experiment with a model is introduced. To solve the problem, the following methods were used: a) expert methods for extraction of initial data; b) mathematical models of private semi-Markov processes; c) methods of transformation of Laplace; r) methods of planning of experiment. Illustrative examples and graphs accompany the task sequence demonstration. As a result of the experiment, the regularities of the studied process, the existence of which was proved formally. As a result of the experiment, the regularities of the studied process were revealed, the existence of which was officially proved. The results of the study broaden knowledge about the application of methods of Markov processes to assess the stability of the functioning of AI elements in relation to the conditions of the impact of threats.

show abstract

Testing the Semi Markov Model Using Monte Carlo Simulation Method for Predicting the Network Traffic

Cited by 3 publications

References 17 publications

HydraGAN A Multi-head, Multi-objective Approach to Synthetic Data Generation

HydraGAN A Multi-head, Multi-objective Approach to Synthetic Data Generation

Markov modeling on dynamic state space for genetic disorders and infectious diseases with mutations: Probabilistic framework, parameter estimation, and applications

Contact Info

Product

Resources

About