System and methods for determining keyboard input in the presence of multiple contact points

Abstract

System and methods are provided for interpreting multiple contact events detected on a virtual keyboard of a computing device to determine an intended key press. The systems and methods utilize event data collected from each of the multiple contact events on a touchscreen in order to provide a general confidence level at which each contact event is deemed to be an intended key press. The confidence level is based on a probabilistic/algorithmic assessment of various criteria for each contact event detected. The criteria may be variable and rely on other events detected on the virtual keyboard even though each contact event is considered a physically independent action from one another. Once an intended key is determined, that key may be input to a word predication application also stored on the computing device. The word prediction application may use the intended key to predict at least one word for the user.