Public | Automated Build

Last pushed: 2 months ago
Short Description
The software engineering project of the fourth semester.
Full Description

wwi15seb_se-project

A repository for the software engineering project (4th semester).

How to use

  • Start the container with the following command: docker run --rm -ti -p 9200:9200 -p 5601:5601 -p 9500:9500 -p 9600:9600 -p 1880:1880 --name container mariusspielberger/se-project
    The startup can take up to three minutes, afterwards it should look somewhat like this:

Double-click on it to open the menu, select the “add new credentials”-option and click on the pencil-button right next to it.

  • Now you’ll need to authenticate with twitter. If you don’t have an account, you must set up one before performing this step. If the authentication has been successful, go back to Node-RED, where you’ll see your Twitter ID. Now press the “add”-button.
  • To finish the Node-RED setup, click on “deploy” in the top right corner.
  • Now, go to Kibana on localhost:5601 or http://192.168.99.100:5601/. In the “discover”-Tab you should be able to see the raw data of the first tweets imported.
  • Click on Dashboards in the panel on the left side. Here you can find pre-configured dashboards visualizing the data.
  • Have fun!

Here-API

In order to provide a heatmap of where tweets have been posted, Kibana needs geohashes. Unfortunately, tweets do not provide this information. This is why we use a free trial of the HERE-API, which provides the geohashes based on location names. In case the 90-day trial runs out we’ll need to configure a new account on https://developer.here.com/?create=Evaluation&keepState=true&step=account and replace the outrun one in Node-RED with it. Take the AppID and the AppCode you receive and enter it into the following node:

Additional Information

Using the snapshot

When you hit deploy in Node-RED, the container starts logging the tweets coming from the twitter API. Unfortunately we're only able to receive the newest tweets (the ones posted after you hit deploy). To provide you a useful dashboard there needs to be a big collection of tweets, so we already included a snapshot with a big collection of tweets in Kibana.
To refer to that Snapshot, navigate into a Dashboard as described in How to use. Click on timestamp in the top right corner and change to absolute time range.

Set the parameters like this:
From: 2017-07-11 11:00:00.000
To: 2017-07-11 14:00:00.000

And press Go to enjoy the data.

Changing the #-Subject

The Tweets you’ll import all refer to the Hashtag “Trump”. In the twitter-node menu (as mentioned in step 3) you have the option to modify the list of Hashtags, for example if you’d rather analyze what people tweet about Putin or puppies.

Other Visualizations

It's also possible to create your own visualizations in Kibana. To do so, navigate into the Visualize-Tab. Click on the + next to the search bar and start creating.

Docker Pull Command
Owner
mariusspielberger

Comments (0)