TU Berlin

Quality and Usability Lab2018_11_05_Kirifidis

Inhalt des Dokuments

zur Navigation

Supporting Learning at Scale with Crowds, Data, and AI

LOCATION:  TEL, Room Auditorium 3 (20th floor), Ernst-Reuter-Platz 7, 10587 Berlin

Date/Time: 05.11.2018, 14:15-15:15 

SPEAKER: Christian Kirifidis (TU Berlin)


The knowledge about the next generations of customers and consumers is changing the way companies act and react on the market. In recent years a new kind of customer has emerged, one that knows its way around technology and has a substantially different understanding of how companies are to provide value to their life. Forrester splits these new Empowered Customers in five segments with varying degrees of empowerment. In the thesis two hypothesis are explored. 

1. The Empowered Customer dimensions can be identified in German Social Media.
2. The detection of Empowered Customer dimensions in German Social Media can be done in an automated way.

To test these hypotheses, a survey with 23 items was developed that aims to identify the five dimensions of empowerment. It was used to crowd label 300 German social media posts related to the Nintendo’s gaming console Switch. Then, the labeled posts, were further processed using text mining techniques to extract features from the texts. Finally, machine learning was used to automatically predict the dimensional values for the posts. Overall this thesis is the first attempt to automatically identify the Empowered Customers based on online posts. 






Schnellnavigation zur Seite über Nummerneingabe