Facial recognition by Facebook and Google is getting better and better. We tried to use this to look at political figures in power and opposition to try and understand posturing.
Our facial emotion analysis of two leaders, during public speeches.
“In politics, when reason and emotion collide, emotion invariably wins” – Drew Westen, The Political Brain
To understand the emotions that Mehbooba Mufti, the chief minister and Omar Abdullah, the leader of opposition, reveal when performing public speeches, and understand the political posturing we looked, with special attention, at their speeches, and how are they saying it, through portrait photos taken during those speeches.
Just like the speeches themselves, politicians take great care with their facial expressions and the message they are conveying. Be it anger, or power posturing, the display of positions can be understood with body language and facial expressions.
We went through 40 recent photos, by date, 20 for Mehbooba Mufti and 20 for Omar Abdullah, from Google Images, portraying the candidates during recent public speeches.
Next, we used Microsoft Cognitive Services API to automatically identify facial attributes related to eight emotions: anger, contempt, disgust, fear, happiness, neutral, sadness, and surprise.
The Emotion API takes a facial expression in an image as an input, and returns the confidence across a set of emotions for each face in the image, as well as bounding box for the face, using the Face API.
And this is what we found:
You can find the data set used in the story here.