I don't believe that we currently have a specific third-party web screen-scraper package that we recommend; we had used a couple in the past for specific customer implementations.
The main point of that chart is - if you have a package with an API (Java), you would use the Java
BizComponent to interface with it, as we had done (I don't currently recall the packages that we used at that time).
For simple cases, you can also use the HTTP BizComp to get the raw HTML, and then parse through the
XML(or text if not well-formed content), without using a screen-scraper.