/swift-tweetSentiment-lab-ios-0616

Primary LanguageSwift

Tweet Sentiment

Let's find out how The Flatiron School is doing on Social Media. We'll need two APIS to do this. One to get the tweets, and another to do what's called sentiment analysis.

Instructions

  1. Open the project and install the STTwitter CocoaPod.
  • STTwitter is written in Objective-C, however, this doesn't mean you'll need to make a bridging header. Having use_frameworks! in the Podfile does some behind the scenes magic for you.

  1. Do the appropriate method call to get all tweets with the query FlatironSchool. Use Twitter "App Only With Consumer Key Authentication". You'll have to get an App Key/Secret here. If you don't have a twitter account already, you'll have to make one.

  2. Once you have the tweets. Send each tweet for sentiment analysis via The Sentiment 140 API. This returns a polarity value of 0 for negative, 2 for neutral and 4 for positive. Technically, we are supposed to use the Bulk Classification Service, but it's a bit confusing. I'd just use the Simple Classification Service. For both the query and the text key you will need to URL encode the String values. You can do this relatively simply with this code:

let unescaped = "Going to campus is always rewarding @FlatironSchool you guys are the best! #LearnLoveCode https://t.co/VGipQWILTu"
let escapedString = unescaped.stringByAddingPercentEncodingWithAllowedCharacters(NSCharacterSet.URLHostAllowedCharacterSet())
print("escapedString: \(escapedString)")
  1. Your App should have just a label. When the app loads, it should request all of the tweets mentioning FlatrionSchool and then give an average sentiment for the term FlatironSchool displayed in the label.