A novel analytical model for assessing mean response time in performance evaluation of cloud server farms using hidden Markov model to improve the accuracy and comparing with SVM.

Autor:	Rajan, Sofia, Parthipan, V.
Předmět:	HIDDEN Markov models MARKOV processes T-test (Statistics) SUPPORT vector machines REACTION time
Zdroj:	AIP Conference Proceedings; 2023, Vol. 2821 Issue 1, p1-8, 8p
Abstrakt:	The aim is to compare the mean response time in performance evaluation of novel cloud server farms using the Hidden Markov model and comparing with the Support Vector Machine model. Mean response time of perfor mance evaluation is performed with two groups: Hidden Markov model and the SVM model of sample size N=10. G power 80% threshold 0.05%, CI 95%. Mean and Standard deviation. Independent sample T-Test was carried out using Hidden Markov model and SVM model. Hidden Markov model (HMM) (92.1%) performs better than SVM (89.6%). There is a statistically significant difference between HMM and SVM with p=0.000, p<0.05 (2-tailed). The results conclude that the proposed Hidden Markov model algorithm has a better mean response time. [ABSTRACT FROM AUTHOR]
Databáze:	Complementary Index
Externí odkaz:	Zobrazit plný text záznamu