Multiscale Entropy Analysis by Using Wiki Page Views on Simplifying User Behavior - Anqing Hu

Recently I'm doing my research on multiscale entropy (MSE), although the concept of entropy was originally a thermodynamic construct, it has been adapted in other fields of study, including ecological economics, information theory, psychodynamics, and so on. Since introduced in early 2000, MSE has found many applications in biosignal analysis and been extended to multivariate MSE. 

Wiki provides us with a huge amount of free complex data could be used as analytic analyses about MSE and multivariate MSE with developed useful analyzing tools. Personally thinking, sharing free complex data makes more people addicted to Wiki. 

By analyzing web pages view, kind of modern biosignals, we can easily see a country's intercultural communication, such as analyzing small languages family page view, see some of a native language's knowledge and information couldn't be translated to another then got a quick glance at the local, or picture the English/Spanish/French speakers' characteristics. Just see my examples without applying MSE but still representativeness as follows,

It seemed like Italians are fond of watching TV? As for me, I haven't searched for a channel before. And why the edits and editors are soo less? Normally it's strange a popular page with few edits.
https://tools.wmflabs.org/topviews/?project=it.wikipedia.org&platform=all-access&date=last-month&excludes=


Figure 1. Topviews Analysis of it.wikipedia.org

The Korean star band "EXO(엑소)" is such a popular team, I can't imagine America or Europe athletic-looking young man watching their shows.[No offense] Or the reason is their team name is in English, most fans know "EXO" instead of how to type "엑소", additionally, 3-letters is easier than "엑소" for a Korean to type.
https://tools.wmflabs.org/topviews/?project=en.wikipedia.org&platform=all-access&date=last-month&excludes=


Figure 2. Topviews Analysis of en.wikipedia.org

There is no doubt that data is the easiest and efficient way to understand a culture. Further, those complex data have no limitations by using MSE to analyze Wiki data and especially combining with web.archive. We may base scientific predictions on data, for instance, use a company's data to forecast stock-market. Whereas positive feedback tends to lead to instability via exponential growth, oscillation or chaotic behavior, negative feedback generally promotes stability, delayed feedback such as page view can be very stable, accurate, and responsive. I'm still considering my proposal and a way to group data. You are welcome to put forward your views.

* Since those ideas are related to my research, reproducing, distributing, displaying or performing a work, or to make derivative works... without my permission are prohibited. Thanks for your cooperation.
* As for Figure 1&2, considering screenshots with links are friendly than Excel Table to show on Blog, I printed my screen as the figure.

Comments