DiscoverText is a cloud-based service that allows researchers to import data from the web, such as tweets, Facebook posts, RSS feeds, and Congressional bills, as well as .zip, Excel, and other files from your desktop, in order to conduct a textual analysis. DiscoverText can be used to securely archive data (the company was developed out of work with the federal government), as well as code, report and validate textual analysis.
Coding can be assigned or shared among a team (managed through a credential system—registered users can be assigned different levels of access), with a very quick and easy coding interface. A customizable machine-learning classifier is currently in beta testing as well, which will allow computer-assisted coding of text. On top of its coding features, DiscoverText also functions as a search engine, a de-duplication and near-duplication clustering engine, and as a repository of metadata.
This feature-rich and quickly evolving tool is worth a look for researchers interested in textual analysis of web-based text and interaction. DiscoverText offers a free 30-day limited trial account (note that you must have and use a Facebook account to create your account in order to scrape Facebook posts). If you think DiscoverText would be a useful research tool, please contact us. With enough interest, we might get a license for DePaul.
DiscoverText Video Tutorials
Webinars are announced on the DiscoverText homepage. These sessions are extremely helpful in illustrating the ease of use and research potential of the service and how much continual behind-the-scenes work goes into improving it on a daily basis.