That kinda sucks, especially in this case, where you get "<" and ">" as values as well as node wrappers.
If you replace "<" and ">" universally, then use DOM methods on the resulting XML, you will get an error when you try to get the value from "Commercial.Name". I suppose this could then be an indication that you need to re-format the "Commercial.Name" value by escaping any of the "bad" entities within the node's value, in this case by exchanging "<" and ">" for "<" and ">"...
Seems like a lot of work. I suppose you could "brute force" the replacement, but not before you replace all of the "<" and ">". Then you could pick apart the XML using JavaScript, taking what you want from the appropriate place within the string.
Either way, your work is cut out for you.