Supervised Classification for Link Prediction in Facebook Ego Networks With Anonymized Profile Information

Riccardo Giubilei () and Pierpaolo Brutti ()
Additional contact information
Riccardo Giubilei: Sapienza University of Rome
Pierpaolo Brutti: Sapienza University of Rome

Journal of Classification, 2022, vol. 39, issue 2, No 6, 302-325

Abstract: Abstract Social networks are very dynamic objects where nodes and links are continuously added or removed. Hence, an important but challenging task is link prediction, that is, to predict the likelihood of a future association between any two nodes. We use a classification approach to perform link prediction on data retrieved from Facebook in the typical form of ego networks. In addition to the more traditional topological features, we also consider the attributes of the nodes—i.e., users’ publicly available profile information—to fully assess the similarity between nodes. We propose two new attribute-based features, validating their predictive power through an extensive comparison with natural competitors from the literature. Finally, one of the proposed features is selected when building a state-of-the-art procedure for link prediction that achieves an average AUROC of 96.59% over 85 test ego networks. Valuable insights on the interpretation of the results in the specific context of friendship recommendation in Facebook are also provided.

Keywords: Link prediction; Classification; Homophily; Supervised learning; Social networks; Facebook (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://link.springer.com/10.1007/s00357-021-09408-2 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:jclass:v:39:y:2022:i:2:d:10.1007_s00357-021-09408-2

Ordering information: This journal article can be ordered from
http://www.springer. ... hods/journal/357/PS2

DOI: 10.1007/s00357-021-09408-2

Access Statistics for this article

Journal of Classification is currently edited by Douglas Steinley

More articles in Journal of Classification from Springer, The Classification Society
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().