KaiBot3000 / slackometer

Webapp using D3 and sentiment analysis to visualize Slack communications.

####Slack-o-meter allows rapid visual analysis of traffic and mood across channels within a Slack team.

The Tech

• [Slack] - A messaging app for teams. Using OAuth, users can allow Slack-o-meter to access their account. With a series of API calls, Slack-o-meter gets the name of the authorized team, a list of the channels, and the recent history of each channel.

• [Sentiment140] - Public sentiment analysis optimized for Twitter, created by Stanford Graduate students. Given a series of comments via JSON, the Sentiment140 bulk classifier returns a sentiment rating from 0-4 for each, with 0 being very negative and 4 being very positive.

• [D3] - Library with a huge variety of graphs and charts. Slack-o-meter uses a pack layout, a series of bubbles with color corresponding to sentiment and size to traffic over the last week. Data from Slack is processed into a Python dictionary, then passed into D3 using JSON. D3 generates svg elements with unique attributes for each item in the JSON.

Installation

In order to run Slack-o-meter on your computer, you'll need to make a Slack app which redirects to "http://127.0.0.1:5000/slacked", and generate a client key and secret. This can be done in a few minutes through Slack's API portal.

Clone repo:

$ git clone https://github.com/KaiDalgleish/slackometer.git slackometer
$ cd slackometer

Install dependencies:

$ pip install -r requirements.txt

Source secrets to your environment:

$ export CLIENT_ID=[Your client id here, for Slack]
$ export CLIENT_SECRET=[Your secret here, for Slack]
$ export MYEMAIL=[Your email here, for Sentiment140]

Run Slackometer server:

$ python server.py

View in your browser, probably at http://127.0.0.1:5000/

Using Slack-o-meter

• Sign in to Slack using OAuth by clicking "Get Team"

• Wait for the APIs to do their magic... then see a visual representation of your team! Color corresponds to average sentiment, while size corresponds to traffic. Hover to see the name of each channel.

Issues?

Slack-o-meter is a work in progress, and your input is welcome!!