The user specifies an input document collection and can immediately begin to process them manually. For each document loaded, the "Output Form" is filled automatically, as much as possible, with extracted values. Locations of these values are highlighted in the “Text Input” area. The user reviews system generated values, revises and updates them if necessary before moving onto the next document. In between documents, the system learns features of correct and incorrect answers by comparing generated values and manually revised ones, and updates the decision model accordingly. Through this revision process, the amount of information that the system is able to automatically prefill correctly grows over time; the “Output Form” for the first few documents may be empty.