Using unsupervised machine learning to determine social networking user groups

Alan Peslak, Wendy Ceccucci, Scott Hunsinger

Research output: Contribution to journalArticlepeer-review

Abstract

The growth of social media and networking has been exponential. From its humble beginnings in 1995 with Classmates.com through the founding of Friendster in 2002, LinkedIn and MySpace in 2003 and Facebook in 2004, social networking has grown to a worldwide phenomenon with nearly 2.89 billion worldwide active users of Facebook alone (Statista, 2022). The number of major social media sites has also grown, though the top active sites in the United States represent most of the social media activity. We examine a 2019 Pew Internet dataset via Unsupervised Machine Learning with the goal of finding Social Networking User Groups. Usage of the top social media websites is combined with relevant demographic and sociographic data to develop three specific clusters of users of social media in the US. Implications for marketers, researchers and society are discussed.

Original languageEnglish (US)
Pages (from-to)215-230
Number of pages16
JournalIssues in Information Systems
Volume23
Issue number2
DOIs
StatePublished - 2022

All Science Journal Classification (ASJC) codes

  • General Business, Management and Accounting

Fingerprint

Dive into the research topics of 'Using unsupervised machine learning to determine social networking user groups'. Together they form a unique fingerprint.

Cite this