The detection and effect of social events on Wikipedia data-set for studying human preferences

Julien Assuied; Yérali Gandica

doi:10.3389/fdata.2023.1077318

The detection and effect of social events on Wikipedia data-set for studying human preferences

Front Big Data. 2023 Mar 30:6:1077318. doi: 10.3389/fdata.2023.1077318. eCollection 2023.

Authors

Julien Assuied¹, Yérali Gandica^{2

3}

Affiliations

¹ CY-Tech Cergy Paris Université, Cergy, France.
² Department of Mathematics & Master in Big Data, Universidad Internacional de Valencia (VIU), Valencia, Spain.
³ CY Cergy Paris Universite, CNRS, Laboratoire de Physique Théorique et Modelisation, Cergy, France.

Abstract

Several studies have used Wikipedia (WP) data-set to analyse worldwide human preferences by languages. However, those studies could suffer from bias related to exceptional social circumstances. Any massive event promoting exceptional editions of WP can be defined as a source of bias. In this article, we follow a procedure for detecting outliers. Our study is based on 12 languages and 13 different categories. Our methodology defines a parameter, which is language-dependent instead of being externally fixed. We also study the presence of human cyclic behavior to evaluate apparent outliers. After our analysis, we found that the outliers in our data-set do not significantly affect the analysis of preferences by categories among different WP languages. While investigating the possibility of bias related to exceptional social circumstances is always a safe measure before doing any analysis on Big Data, we found that in the case of the first ten years of the Wikipedia data-set, outliers do not significantly affect using Wikipedia data-set as a digital footprint to analyse worldwide human preferences.

Keywords: Wikipedia; human preferences; massive events; outliers detection; possible bias.

Grants and funding

This work was supported by the OpLaDyn grant obtained in the 4th round of the Trans-Atlantic Platform Digging into Data Challenge (2016-147 ANR OPLADYN TAP-DD2016).