This shows the system that I have been working on. It extracts the data from an API feed, parses it and cleans it, and then feeds it into Protege, at which point a case-based reasoning system is built around the specifics. While it currently extracts web products and compares on a collection of critical data categories, the system can be created with any structured or semi-structured data set. The more structured, the easier the data extraction process – the data must be immaculate in order to go into the case-based reasoning system.
I don’t know why I called it the Drunken Master, except that it sounds like a cool code-name. If anyone is interested in working together on this, please feel free to write me!
- Hadoop: The Definitive Guide (shop.oreilly.com)
- Make Case-Based Approximate Reasoning a Reality (arnoldit.com)
- The ‘Drunken Master’: A Prototype Case-Based Reasoning System (velluminformation.com)