User input prediction

A method comprising:

• outputting, by a computing device comprising one or more processors, a plurality of targets via a user interface, the plurality of targets representing respective user-selectable items of information;
• receiving, by the computing device, data indicative of a user input event, the user input event being entered in the user interface via at least one location of the plurality of targets;
• and
• in response to the user input event, identifying, by the computing device, a probable target of the plurality of targets, by modeling the user input event with a trained model;
• wherein the trained model corresponds to the probable target, wherein the trained model is configured to model locations of previous input events received in the user interface for the probable target, and wherein the trained model is used based on a likelihood of observing data indicative of the user input event for the probable target.

A non-transitory computer-readable storage medium, comprising computer-executable instructions, wherein the instructions, when executed by a computing device, cause the computing device to perform operations comprising:

• outputting a plurality of targets via a user interface, the plurality of targets representing respective user-selectable items of information;
• receiving data indicative of a user input event, the user input event being entered in the user interface via at least one location of the plurality of targets;
• and
• in response to the user input event, identifying, by the computing device, a probable target of the plurality of targets, by modeling the user input event with a trained model;
• wherein the trained model corresponds to the probable target, wherein the trained model is configured to model locations of previous input events received in the user interface for the probable target, and wherein the trained model is used based on a likelihood of observing data indicative of the user input event for the probable target.