A machine-learning apprentice for the completion of repetitive forms

Abstract

A non-intrusive assistant or apprentice is described that provides viable default values for blank fields in a form and highlights possible inconsistencies in completed fields. The focus is on the intermediate processing stage of form-filling, so the existence of an electronically reproduced on-screen form is assumed. The learning apprentice can be used for repetitious form-filling tasks. The authors establish the need for learning apprentices, describe a framework for such a system, and explain the difficulties of form-filling. Empirical results show that the system significantly reduced the number of keystrokes required to fill out repetitious forms by 87%, and correctly predicted nearly 90% of the values on the form.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>