Election 2016: The big data showdown - Google I/O 2016

Опубликовано 20 мая 2016, 21:37

The US election is still months away, but a massive amount of data has already been generated on the topic. From social media to campaign finance disclosures to newspaper articles to poll results, we are awash in data about the candidates. How can we separate the signal from the noise, and figure out what it all means?

This talk takes a number of disparate sources of data; reddit comments, federal election disclosures, and a global event database as the raw materials to use Dataflow, Datalab, Cloud ML, and BigQuery, to extract meaning from it all. We find out what people are saying, in real-time, about the candidates. We show how to build a dashboard that can display front-runners in the campaign donation race, the reddit popularity contest, and more.

Learn how to set up BigQuery and re:dash to analyze the elections: goo.gl/VFWJKV

You'll see how you, too, can gain insight from disparate data sources by harnessing the suite of tools in Google's cloud.

See all the talks from Google I/O 2016 here: goo.gl/olw6kV
Watch more Android talks at I/O 2016 here: goo.gl/Uv3jls
Watch more Chrome talks at I/O 2016 here: goo.gl/JoMLpB
Watch more Firebase talks at I/O 2016 here: goo.gl/JTH9Fr
Subscribe to the Google Developers Channel: goo.gl/mQyv5L

#io16 #GoogleIO #GoogleIO2016