For this week’s homework in Designing for Data Personalization with Sam Slover, I made progress on a project that I’m working on for Fusion as part of their 2016 US Presidential Election coverage. I began this project by downloading all the images from each candidate’s Twitter, Facebook, and Instagram account — about 60,000 in total — then running those images through Clarifai‘s convolutional neural networks to generate descriptive tags.

With all the images hosted on Amazon s3, and the tag data hosted on parse.com, I created a simple page where users can explore the candidates’ images by topic and by candidate. The default is all topics and all candidates, but users can narrow the selection of images displayed by making multiple selections from each field. Additionally, more images will load as you scroll down the page.

Screen Shot 2015-09-17 at 11.24.47 AM

Screen Shot 2015-09-17 at 11.30.30 AM

Screen Shot 2015-09-17 at 11.20.50 AM

Screen Shot 2015-09-17 at 11.28.45 AM

Screen Shot 2015-09-17 at 11.25.42 AM

Screen Shot 2015-09-17 at 11.19.53 AM

Unfortunately, the AI-enabled image tagging doesn’t always work as well as one might hope.

Screen Shot 2015-09-17 at 11.23.49 AM

Here’s the page’s JavaScript code: