Open topic with navigation
A binary category is a special kind of category, designed to answer yes/no questions about documents, files, and text, like Is this document spam?, Does this violate company policy?, or Is this work-related?.
After you create a binary category, you train it. Unlike regular categories, which receive only positive training, binary categories can receive both positive and negative training. You provide the binary category with documents or text that result in a yes (POSITIVE) answer when querying with the binary category. You also provide documents or text that result in a no (NEGATIVE) answer to the same query.
After training, the binary category determines whether a piece of text, a file on disk, or a document in IDOL results in a POSITIVE or a NEGATIVE result. IDOL also provides a score (0–1) for the result, which indicates the confidence.