Research talk:Measuring article importance/Work log/2014-10-28

From Meta, a Wikimedia project coordination wiki

Tuesday, October 28, 2014[edit]

Today, I'm doing a comparison between different measures of importance.

First, let's replicate Nettrom's work by looking at the distribution of pageview rates per WikiProject importance classification. In this document, I'll take the ROUND(AVG(Importance class)) per article as the standardized importance class. I've tested MIN(Importance class) and MAX(Importance class) and found no substantial differences.

First, let's looks at the relationship between view rate and importance.

The density of log(views) is plotted for English Wikipedia articles by the avg. WikiProject importance classification.
View density by importance. The density of log(views) is plotted for English Wikipedia articles by the avg. WikiProject importance classification.
The density of log(views) is plotted for English Wikipedia articles by the avg. WikiProject importance classification. (zero values removed)
View density by importance (bounded). The density of log(views) is plotted for English Wikipedia articles by the avg. WikiProject importance classification. (zero values removed)

Well, that's surprising. I expected this to look like Morten's graph. It looks like there are a huge number of articles that don't receive any views.

Next, let's look at the relationship between inlinks and importance.

The density of log(# of inlinks) is plotted for English Wikipedia articles by the avg. WikiProject importance classification.
Inlink density by importance. The density of log(# of inlinks) is plotted for English Wikipedia articles by the avg. WikiProject importance classification.

Now that looks like I'd expect (though admittedly with more error than I'd like).


Now how about we look at the relationship between views and inlinks together.

The density of log(# of inlinks) and log(views) is plotted for English Wikipedia articles by the avg. WikiProject importance classification.
Inlink and view density by importance. The density of log(# of inlinks) and log(views) is plotted for English Wikipedia articles by the avg. WikiProject importance classification.

OK. Time to talk to User:Nettrom about my weird results with page views. --Halfak (WMF) (talk) 21:29, 28 October 2014 (UTC)[reply]