IDENTIFYING DECEPTIVE SOCIAL MEDIA CONTENT

Approaches presented herein enable identifying a deceptive social media post such as a fraudulent social survey in a social media environment. Specifically, a regression model including one or more factors, each of the one or more factors includes a respective parameter, is generated based on an ana...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Grant, Robert H, Denner, Gary, Hewitt, Trudy L, Dunne, Jonathan
Format: Patent
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Approaches presented herein enable identifying a deceptive social media post such as a fraudulent social survey in a social media environment. Specifically, a regression model including one or more factors, each of the one or more factors includes a respective parameter, is generated based on an analysis of prior social media scam data. A predictor function including a threshold value is generated based on the regression model. A repudiation value for a particular social media post is calculated using the predictor function based on one or more factors and respective parameters. If the repudiation value exceeds the threshold value, a repudiation is provided indicating the social media post is potentially fraudulent.