Multiscale Entropy Analysis of Page Views: A Case Study of Wikipedia
Open Access
Abstract
In this study, the Wikipedia page views for four selected topics, namely, education, the economy/finance, medicine, and nature/environment from 2016–2018 are collected and the sample entropies of the three years’ page views are estimated and investigated using a short-time series multiscale entropy (sMSE) algorithm for a comprehensible understanding of the complexity of human website searching activities. The sample entropies of the selected topics are found to exhibit different temporal variations. In the past three years, the temporal characteristics of the sample entropies are vividly revealed, and the sample entropies of the selected topics follow the same tendencies and can be quantitatively ranked. By taking the 95% confidence interval into account, the temporal variations of sample entropies are further validated by statistical analysis (non-parametric), including the Wilcoxon signed-rank test and the Mann-Whitney U-test. The results suggest that the sample entropies estimated by the sMSE algorithm are feasible for analyzing the temporal variations of complexity for certain topics, whereas the regular variations of estimated sample entropies of different selected topics can’t simply be accepted as is. Potential explanations and paths in forthcoming studies are also described and discussed.Keywords
This publication has 25 references indexed in Scilit:
- Weighted multivariate composite multiscale sample entropy analysis for the complexity of nonlinear times seriesPhysica A: Statistical Mechanics and its Applications, 2018
- Symbolic phase transfer entropy method and its applicationCommunications in Nonlinear Science and Numerical Simulation, 2017
- Sample Entropy, Univariate, and Multivariate Multi-Scale Entropy in Comparison with Classical Postural Sway Parameters in Young Healthy AdultsFrontiers in Human Neuroscience, 2017
- Multiscale Symbolic Phase Transfer Entropy in Financial Time Series ClassificationFluctuation and Noise Letters, 2017
- Big Data: A Revolution That Will Transform How We Live, Work, and ThinkInternational Journal of Advertising, 2014
- Early Prediction of Movie Box Office Success Based on Wikipedia Activity Big DataPLOS ONE, 2013
- Multiscale entropy analysis of biological signalsPhysical Review E, 2005
- Multiscale Entropy Analysis of Complex Physiologic Time SeriesPhysical Review Letters, 2002
- What is physiologic complexity and how does it change with aging and disease?Neurobiology of Aging, 2002
- Physiological time-series analysis using approximate entropy and sample entropyAmerican Journal of Physiology-Heart and Circulatory Physiology, 2000