An Improved Efficiency in Envisioning the Personage Traits over Online Social Media based on Indian Metrics during Pandemic using Novel Naive Bayes Classifier Algorithm Comparing with Logistic Regression Algorithm

Aim: The primary aim of this research is to increase the intensity percentage of personage traits detection to reveal the impact of coronavirus on Twitter users by utilizing machine learning classifier algorithms by comparing Novel Naive Bayes Classifier algorithm and Logistic Regression algorithm. Material(s) and Method(s): Naive Bayes Classifier algorithm with test size=10 and Logistic Regression algorithm with test size=10 was estimated several times to envision the efficiency percentage with confidence interval of 95% and G-power (value=0.8). Naive Bayes classifier compares whether a specific feature in a class is unrelated to another feature. A logistic regression model predicts the probability of an item belonging to one group or another. Results and Discussion: Naive Bayes algorithm has greater efficiency (86%) when compared to Logistic Regression efficiency (60%). The results achieved with significance value p=0.169 (p>0.05) shows that two groups are statistically insignificant. Conclusion(s): Naive Bayes Algorithm executes remarkably greater than the Logistic Regression algorithm. Copyright © 2022 Wolters Kluwer Medknow Publications. All rights reserved.

An Improved Efficiency in Envisioning the Personage Traits over Online Social Media based on Indian Metrics during Pandemic using Novel Naive Bayes Classifier Algorithm Comparing with Logistic Regression Algorithm

Aim: The primary aim of this research is to increase the intensity percentage of personage traits detection to reveal the impact of coronavirus on Twitter users by utilizing machine learning classifier algorithms by comparing Novel Naive Bayes Classifier algorithm and Logistic Regression algorithm. Materials and Methods: Naive Bayes Classifier algorithm with test size=10 and Logistic Regression algorithm with test size=10 was estimated several times to envision the efficiency percentage with confidence interval of 95% and G-power (value=0.8). Naive Bayes classifier compares whether a specific feature in a class is unrelated to another feature. A logistic regression model predicts the probability of an item belonging to one group or another. Results and Discussion: Naive Bayes algorithm has greater efficiency (86%) when compared to Logistic Regression efficiency (60%). The results achieved with significance value p=0.169 (p>0.05) shows that two groups are statistically insignificant. Conclusion: Naive Bayes Algorithm executes remarkably greater than the Logistic Regression algorithm.