Social media metrics and sentiment analysis to evaluate the effectiveness of social media posts

Autor(en)
Flora Poecze, Claus Ebster, Christine Strauss
Abstrakt

The present paper presents the results of an analysis of indicators underlying successful self-marketing techniques on social media. The participants included YouTube gamers. We focus on the content of their communication on Facebook to identify significant differences in terms of their user-generated Facebook metrics and commentary sentiments. Methodologically, ANOVA and sentiment analysis were applied. ANOVA of the classified post categories revealed that re-posted YouTube videos gained significantly fewer likes, comments, and shares from the audience. On the other hand, photos tended to show significantly more follower-generated actions compared to other post types in the sample. Sentiment analysis revealed underlying follower negativity when user-generated activity tended to be relatively low, offering valuable complementary results to the mere analysis of other post indicators, such as the number of likes, comments, and shares. The results indicated the necessity to utilize natural language processing techniques to optimize brand communication on social media and highlighted the importance of considering the opinion of the masses for better understanding of consumer feedback.

Organisation(en)
Institut für Rechnungswesen, Innovation und Strategie
Externe Organisation(en)
Fachhochschule Burgenland
Journal
Procedia Computer Science
Band
130
Seiten
660-666
Anzahl der Seiten
7
ISSN
1877-0509
DOI
https://doi.org/10.1016/j.procs.2018.04.117
Publikationsdatum
04-2018
Peer-reviewed
Ja
ÖFOS 2012
502050 Wirtschaftsinformatik, 502007 E-Commerce
Schlagwörter
ASJC Scopus Sachgebiete
Allgemeine Computerwissenschaft
Link zum Portal
https://ucrisportal.univie.ac.at/de/publications/645f7593-38fc-4f87-baae-686836f48ed8