Follow the instructions below to export a dataset so you can archive the data or analyze it in another application.
It is not possible to export data from an archive in DiscoverText. To download any Twitter data (from either Gnip or the public Twitter API), you need to download the data from a dataset instead. This is to comply with Twitter's terms of service (they require that any data be processed in some form before an export is allowed).
- Create a dataset if you haven't already.
- Open the dataset.
- In the Dataset Options section, click Export dataset.
- Select the check boxes for the desired options.
- Include unit text: The full text of each item in the dataset.
- Include metadata: The attributes associated with each item.
- Include codes and coder: The codes assigned by the coder.
- Include classification: The specific code classification for each item.
- Include reference text: The memos and bookmarks as well as any other reference information that is available.
- Include adjudications: The adjudications and results of the process
- Include annotations: The annotations added to each item.
- Include attachments: The files attacted to the datasets.
- From the Output format, select the file type. Note: Under Twitter's Terms of Service, this must be CSV File. Twitter data cannot be saved as XML or JSON, and is limited to no more than 50,000 tweets per day.
- Click Export.
- If exporting Twitter data, you will be reminded that credits will deducted from account. Click OK to continue with the export.
- On the Exports page, refresh the page to see the status of the export.
- When the export is complete, click download next to the file name.
The file is available to download for seven days.
Exported datasets are available through the left navigation:
- On the left navigation, click on Analytics / Export.
- Click on Exported Files.