TU Berlin

Quality and Usability Lab2017_07_03_Kuester

Page Content

to Navigation

There is no English translation for this web page.

Predicting personality traits from touch-screen based interaction

LOCATION:  TEL, Auditorium 1 (20th floor), Ernst-Reuter-Platz 7, 10587 Berlin

Date/Time: 09.08.2017, 12:00-12:30

SPEAKER: Ludwig Küster (TU Berlin)


The prediction of personality from behavioural data has drawn substantial attention in recent years. Many approaches rely on comprehensive long-term data sets. In this talk results will be presented that data from four minutes of touch-screen interaction can be used to predict a user’s ‘extraversion’ and ‘openness’ with 71% accuracy. Interaction records from the PflegeTab application were mapped on to the big-5 personality dimensions of 75 subjects. Prediction results were compared between linear-regression, knn, random-forest and SVM.


Quick Access

Schnellnavigation zur Seite über Nummerneingabe