GDPR-Compliant Social Network Link Prediction in a Graph DBMS: The Case of Know-How Development at Beekeeper
In: Knowledge, Jg. 2 (2022-05-19), S. 286-309
Online
unknown
Zugriff:
The amount of available information in the digital world contains massive amounts of data, far more than people can consume. Beekeeper AG provides a GDPR-compliant platform for frontline employees, who typically do not have permanent access to digital information. Finding relevant information to perform their job requires efficient filtering principles to reduce the time spent on searching, thus saving work hours. However, with GDPR, it is not always possible to observe user identification and content. Therefore, this paper proposes link prediction in a graph structure as an alternative to presenting the information based on GDPR data. In this study, the research of user interaction data in a graph database was compared with graph machine learning algorithms for extracting and predicting network patterns among the users. The results showed that although the accuracy of the models was below expectations, the know-how developed during the process could generate valuable technical and business insights for Beekeeper AG.
Titel: |
GDPR-Compliant Social Network Link Prediction in a Graph DBMS: The Case of Know-How Development at Beekeeper
|
---|---|
Autor/in / Beteiligte Person: | Korányi, Rita ; Mancera, José A. ; Kaufmann, Michael |
Link: | |
Zeitschrift: | Knowledge, Jg. 2 (2022-05-19), S. 286-309 |
Veröffentlichung: | MDPI AG, 2022 |
Medientyp: | unknown |
ISSN: | 2673-9585 (print) |
DOI: | 10.3390/knowledge2020017 |
Schlagwort: |
|
Sonstiges: |
|