[Paper Review] Modelling page-view dynamics on Wikipedia

Yana Volkovich, David Laniado+1 more|arXiv (Cornell University)|Jan 1, 2013

Wikis in Education and Collaboration14 references3 citations

TL;DR

This paper proposes a Poisson process-based model to predict page-view dynamics of promoted Wikipedia content, showing that popularity decays exponentially over time when circadian user activity patterns are accounted for. The model offers a simpler, empirically validated explanation for content popularity evolution than prior approaches.

ABSTRACT

We introduce a model for predicting page-view dynamics of promoted content. The regularity of the content promotion process on Wikipedia provides excellent experimental conditions which favour detailed modelling. We show that the popularity of an article featured on Wikipedia’s main page decays exponentially in time if the circadian cycles of the users are taken into account. Our model can be explained as the result of individual Poisson processes and is validated through empirical measurements. It provides a simpler explanation for the evolution of content popularity than previous studies.

Motivation & Objective

• To model the temporal dynamics of page views for content promoted on Wikipedia’s main page.
• To understand how user activity cycles—particularly circadian rhythms—affect the decay of article popularity.
• To develop a parsimonious model that explains popularity evolution more simply than previous studies.
• To validate the model using empirical page-view data from Wikipedia.

Proposed method

• Modeling individual user interactions as Poisson processes to capture stochastic user behavior.
• Incorporating circadian user activity cycles into the model to reflect time-of-day variations in page views.
• Deriving a time-dependent decay function for article popularity based on aggregated Poisson processes.
• Fitting the model to real-world Wikipedia page-view data to validate its predictive accuracy.
• Using empirical measurements to confirm the exponential decay of popularity under the model’s assumptions.
• Comparing model predictions with observed data to assess goodness of fit and explanatory power.

Experimental results

Research questions

• RQ1 How does the popularity of a Wikipedia article promoted on the main page evolve over time?
• RQ2 To what extent do circadian user activity patterns influence the decay rate of page views?
• RQ3 Can a Poisson process model accurately describe the dynamics of page-view counts for promoted content?
• RQ4 How does this model compare in simplicity and accuracy to prior models of content popularity?

Key findings

• The popularity of a promoted Wikipedia article decays exponentially over time when circadian user activity is considered.
• The model, based on individual Poisson processes, provides a simpler explanation for popularity dynamics than previous approaches.
• Empirical data confirms the exponential decay pattern, supporting the model’s validity.
• The inclusion of circadian cycles significantly improves the model’s fit to real-world page-view data.
• The model successfully captures the temporal evolution of page views with minimal assumptions.
• The results suggest that user behavior patterns, particularly daily rhythms, are critical for predicting content popularity.