Caution to reader: this blog post is 3 months or older. Blog posts older than three months may contain details about the Hack Reactor program that are no longer accurate. Please refer to other pages on our website to confirm current information and email us with questions.
When a team of Hack Reactor students won awards from Twitter and Nexmo at the API Con Hackathon, they knew they had an attention grabbing project.
“We got such good reception, we thought it would be worthwhile to expand the project,” says Drew Cuthbertson. Burgeoned by this success, they expanded their idea over several weeks at Hack Reactor. The result is NetSense, a program to monitor and respond to sentiment data produced by people across the world. NetSense was built by Cuthbertson, Ash Hoover, Andrew Krause and Nick Wei.
NetSense retrieves data from Twitter: each tweet produces a time and location stamp and a sentiment score. Sentiment is measured by a human language processor which judges whether the tweet is positive or negative, and can assess several gradients on each side. With this data, NetSense can produce graphs and maps describing local events, like a concert or conference, or global events, like the World Cup:
The ability to observe sentiment across time and space has a broad range of applications.
“You can see what people think of your company,” says Cuthbertson. “You might find that you’re very popular in Los Angeles, but not so much in San Francisco.”
In addition to data monitoring, NetSense has a host of responsive features as well. Users can auto-retweet and auto-reply to tweets based on sentiment. The natural language processing component can be used to set up an automated help desk. The system integrates with email and Google Calendar, so that people can set up events and schedule tweets without working through multiple programs.
While the group used Twitter as a proof of concept, it would not be especially difficult to incorporate data from Facebook and LinkedIn.
NetSense had its first launch in June, and the timing happened to be fortuitous.
“We had our first public release the day before Google I/O,” Cuthbertson recalls. “First day of Google IO they announce Google Cloud Dataflow which is extremely similar to this product., though it doesn't have the automations. I'm just glad we can say we had our first public launch before Google announced.”