Skip to content

linguistic analysis to detect emotional tones in tweets

Notifications You must be signed in to change notification settings

danhendrix/VTAK

 
 

Repository files navigation

Sentiment.ly

Twitter users analyzed, using the brains of IBM Watson's Tone Analyzer.

Wouldn't it be great if there were a web app where you could "test drive" a social media post or text message and have IBM Watson's powerful sentiment analysis tool tell you how that text is likely to be perceived?

Wouldn't you love it if you could run someone's tweets through Watson and find out how their expressed sentiments will ACTUALLY be perceived?

Enter Sentiment.ly!


Our Team

Table of Contents

  1. Usage
  2. Requirements
  3. Development
    1. Installing Dependencies
  4. Team
  5. Contributing

Usage

Analyze a Twitter handle

Entering a valid twitter handle into the corresponding input field will:

  • Return to the user tone analysis from IBM Watson's Sentiment analyzer over a range of 13 different categories by calculating average values of the user's last 50 tweets.
  • Dynamic graphical visualization using d3.
  • Add the search to the database and re-render.

Test drive a tweet

Entering text into the test drive input field will:

  • Analyze the text as if it were a tweet by running it through IBM Watson's sentiment analyzer.
  • Displays dynamic d3 rendering for the individual tweet.

Requirements

  • Node 0.10.x
  • bluebird 3.5.0
  • body-parser" 1.17.1
  • dotenv 4.0.0
  • express 4.15.2
  • mongoose 4.9.0
  • morgan 1.8.1
  • q 1.4.1
  • watson-developer-cloud 2.25.1

Development

Installing Dependencies

From within the root directory:

npm install

Contributing

See CONTRIBUTING.md for contribution guidelines.

See Sentiment.ly tech stack file above, since it apparently doesn't feel like loading here today

About

linguistic analysis to detect emotional tones in tweets

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • JavaScript 54.7%
  • CSS 25.0%
  • HTML 20.3%