Social Media Responsive Robot Week 5 and 6
For week 5 we tried creating some more different prototypes for the environment, but they were not successful, so for week 6 we plan to develop the pop up and cut out environments from week 4 further.
We were successful in connecting the robot’s behaviour to the sentiment of tweets, to display positive, neutral and negative behaviour. For week 6 we plan to link the robot’s behaviour to levels of positivity and levels of negativity. Below is a demonstration of the robot responding to tweets.
Week 6 Aims:
-Design positive, neutral and negative scenes for the environment
-Program levels of robot’s behaviour linked to the levels of sentiment
-Program a GIF generator and make sure the digital system all works
For the environment, we created a background with cardboard and paper, with two states – positive and negative. We were not able to finish all the mechanics of it in time for the project faire, but below are some videos that show what we want to achieve.
Feedback from project faire:
- Research into colour theory. Add lights (Sarah Rubidge)
- Psychology of colour & saturation and intensity of colour rather than the hue. Interaction of Colour(Marina Castan)
- Think about if certain materials express different emotions.(George)
- Robot physical behaviour:
- Make more use of the eye movement. Eyes are the most expressive, and provide psychological insight. (Sarah Rubidge)
- Try experimenting with linking the robot’s points of movement with own movement and see if more natural behaviour linked to different moods can start emerging. Or create different versions of Hex that display variations of certain moods and then have people vote to see which kinds of behaviours people respond to the most. (Chris Leung)
- Sentiment Analysis:
- Look into using neural networks. Word vectors (George)
- Other feedback:
- Create better social media presence for the hex (Alexandra Deschamps-Sonsino)
- Look into Punch Drunk’s children in tech event. (Alexandra Deschamps-Sonsino)
- What if Hex is the third party in a conversation. Comments on the conversations between two other twitter accounts instead of one to one conversation. (Luca Dellatorre)