The
Text Topic node enables you to define, discover, and modify sets of
topics contained in your collection. A topic is a collection of terms
that are strongly associated with a subset of your collection.
Each document can contain zero, one, or many topics. Also, terms that
are used to describe and define one topic can be used in other topics.
There are no rules to determine how many topics you should create.
The number of topics that you create is your preference and will vary
from problem to problem.
The Text
Topic node must be preceded by a Text Parsing node. If it is also
preceded by a Text Filter node, then it will use the term weights
set in that node. Otherwise, the Text Topic node will create term
weights with the Log frequency weighting and Entropy term weighting
settings.