This blog post makes you aware of the Explore example application that I am making available that helps those interested understand how to extract and analyze information from XBRL-based financial reports. Go to the Explore web page to read the documentation, download the database application and Excel information extraction tools, etc.
Read the documentation to get a good overview of what the database application is and does. Or, watch this video walk through.
Essentially, there are 109,778 financial reports from 3,600 public companies that use 17 different US GAAP reporting styles to submit information to the SEC.
This is an excellent starting point (i.e. training data) that can be used to create machine learning applications related to US GAAP financial reporting. I might create a similar set for IFRS. There is lots of VBA code that helps you understand the proper steps of extracting information from XBRL-based reports. Don't make the mistake of focusing on the actual code (i.e. I am not really a very good programmer); focus on the necessary steps to be effective.
Note that if you want to update the database with additional reports, you can use the SEC's RSS feed of reports to do so.
If you cannot open a Microsoft Access database but you want the data, email me and I can get you a copy of the data in Excel.