How 140kit Works
This quick tutorial will provide you with an idea of how to use 140kit for your own work.
The best way to think about this platform is as a storage and analysis engine. When you have an online campaign coming up, are aware of a potentially politically contentious online debate, or have some area of interest for twitter data whatsoever, throw parameters into our system. In our terms, we define a set of twitter data collected on our system, no matter what type of data it is, as a dataset. On your dashboard page, you can create a dataset very easily:
Click on +New Dataset, and you’ll be taken to a page that looks like this:
At this point, you’ve gotta make a decision – what type of data do you want to collect? If you want data surrounding a set of keywords, you want to choose a terms-based approach. If you want to get data from people geo-located within a given region, location will be your method (note that this data is only from people who use geo-locating devices as their tweeting platforms or have otherwise enabled locations to be added to their tweets. This may or may not bias your data, no published work has been done on this question). Otherwise, you can follow a set of users by choosing the user route.
In this example, we choose a term, and our term is “test”. You can append multiple terms together with a comma to search for many terms at once. Be aware that this system is case sensitive, though Twitter’s API is not. This is to allow for very precise control of the data you want, beyond what is provided by Twitter. The next step here is to use the slider to define how long you will collect data (currently, the range is one minute to one week, though this is subject to change in the future).
When you’re ready, create the dataset, confirm the creation, and you’ll be whisked away to the dataset page:
Next, you have to wait until your dataset receives at least one tweet. When this happens, we count this dataset as “valid” and allow you to add analytics from our current offerings. To add them, simply click add analytics:
When you click this, you’ll be presented with a listing of all the analytics that we provide. For total transparency, we link you to an explanation of the analytic, an overview of any and all variables you can manipulate, and a link to the source code. This allows you to get a very clear sense of what will happen when you add the analytic. If you don’t really care, you can just go ahead and click it and see what the results look like in practice.
To add an analytic, simply click “Add to this dataset”, and work your way through the addition page:
This will send a task into our backend, which will be picked up by one of our machines when the dataset has finished streaming and is loaded into the active database. When added, you’ll have a row in the “analytics added” section:
And on the dataset page, you’ll have a row that allows you to view results, which takes you to the analytic’s page. This page is defined by the developer, and will be unique to each analytic. In the case of basic histograms, you will see something like this:
And that’s it! Go have fun!