What I want to do is take unstructured text as found in a book whether fiction ornon-fiction and get a visual that shows topics, clusters passages/sentences according to topic, and relates them semantically.
I have two purposes for this. First, as a writer, to compile notes written over almost two decades so that I can find everything I've written related to a topic of interest. Second, as a reader, to map books and quickly find passages related to a topic of interest.
Can RM do this? Or am I barking up the wrong tree investigatint text analysis tools? Maybe there's a simpler type of software to do this?
Welcome to boards. RapidMiner can do this but it doesn't come as a prepackaged type of system, so you'd have to build the processes to do all of that and possibly hack some R or Python.
Another possible easier way is to investigate using the following 3rd party extensions: Aylien and Rosette. They have support for sentence and entity extraction, taxonmy categorization, and topics.
Another good starting point for text analytics with RapidMiner is also this video series here:
It is a bit outdated but the concepts still hold. And then there of course the 3rd party tools T-Bone has mentioned.
Hope this helps,