Computational Communication Science Conference & LDA for analysing visual frames

Last month it finally happened: The Computational Communication Science workshop conference took place in Hannover! We worked towards this as a team for almost two years. More than 100 participants took part in the discussions and method workshops were held in advance. Apart from some minor glitches, the conference was a complete success. As the main organizer, I wasn’t able to enjoy the conference so much, but the extremely positive feedback from the young and senior scholars makes me very happy in retrospect. I also had time to attend Wouter van Atteveldt’s workshop on Topic Modeling between organizational tasks. The workshop was fantastic and helped me a lot with my project about visual framing of politicians on Facebook. To identify the frames I will now perform a latent dirichlet allocation. At the moment I am still analyzing the images through the Google Vision API. With over 350,000 photo posts, it unfortunately takes some time. But the syntax is surprisingly simple, because with RoogleVision there is already an R packet for this purpose. So my code for this task will be on github shortly.

Schreibe einen Kommentar

Deine E-Mail-Adresse wird nicht veröffentlicht. Erforderliche Felder sind mit * markiert.

Diese Website verwendet Akismet, um Spam zu reduzieren. Erfahre mehr darüber, wie deine Kommentardaten verarbeitet werden.