Method, apparatus, and system for modeling passive and active user interactions with a computer system
First Claim
1. A method for modeling user activity with regard to a plurality of different software applications on a computing device, the method comprising, with the computing device:
- generating semantic descriptions for each of a plurality of user interactions involving the plurality of different software applications over time, wherein (i) at least a first one of the semantic descriptions relates to gaze-tracking data comprising one or more real-time sensor inputs indicative of a user'"'"'s gaze in relation to a display of the computing device during a user interaction involving one of the plurality of different software applications; and
(ii) at least a second one of the semantic descriptions relates to application event data generated as a result of another interaction involving another one of the plurality of different software applications;
algorithmically drawing an inference about a current user activity with the computing device based on at least (i) the first one of the semantic descriptions relating to the gaze-tracking data obtained during the user interaction involving the one of the plurality of different software applications and (ii) the second one of the semantic descriptions relating to the application event data generated as a result of the other interaction involving the other one of the plurality of different software applications and (iii) a recency or frequency of user attention to a user interface element;
developing a model of user activity in relation to the computing device based on associations of the semantic descriptions with the gaze-tracking data over time, initially storing the associations in a short-term memory portion of the model, and moving at least a portion of the associations to a long-term memory portion of the model after a period of time; and
making the inference available for use by a computer-executable module of the computing device.
1 Assignment
0 Petitions
Accused Products
Abstract
A method, apparatus, and system for modeling user interactions with a computer system associates semantic descriptions of passive and active user interactions, which are meaningful at a user level, with application events and user interaction data as a user interacts with one or multiple software applications with a computing device, and uses those associations to build and maintain a user-specific contextual model. In some embodiments, the contextual models of multiple users are leveraged to form one or more collective contextual user models. Such models are useful in many different applications.
32 Citations
26 Claims
-
1. A method for modeling user activity with regard to a plurality of different software applications on a computing device, the method comprising, with the computing device:
-
generating semantic descriptions for each of a plurality of user interactions involving the plurality of different software applications over time, wherein (i) at least a first one of the semantic descriptions relates to gaze-tracking data comprising one or more real-time sensor inputs indicative of a user'"'"'s gaze in relation to a display of the computing device during a user interaction involving one of the plurality of different software applications; and
(ii) at least a second one of the semantic descriptions relates to application event data generated as a result of another interaction involving another one of the plurality of different software applications;algorithmically drawing an inference about a current user activity with the computing device based on at least (i) the first one of the semantic descriptions relating to the gaze-tracking data obtained during the user interaction involving the one of the plurality of different software applications and (ii) the second one of the semantic descriptions relating to the application event data generated as a result of the other interaction involving the other one of the plurality of different software applications and (iii) a recency or frequency of user attention to a user interface element; developing a model of user activity in relation to the computing device based on associations of the semantic descriptions with the gaze-tracking data over time, initially storing the associations in a short-term memory portion of the model, and moving at least a portion of the associations to a long-term memory portion of the model after a period of time; and making the inference available for use by a computer-executable module of the computing device. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17)
-
-
18. A computing system to develop a semantic model of user attention to user interface elements of a plurality of software applications, the computing system comprising:
-
a display; a sensor subsystem to obtain gaze-tracking data, the gaze-tracking data being indicative of a user'"'"'s gaze in relation to the display; a framework embodied in one or more non-transitory machine-accessible media, the framework configured to, over time, during a plurality of user interactions involving a plurality of different software applications, generate semantic descriptions for each of the user interactions, wherein (i) at least a first one of the semantic descriptions relates to locations on the display corresponding to the gaze-tracking data; and
(ii) at least a second one of the semantic descriptions relates to application event data;a model embodied in the one or more non-transitory machine-accessible media, the model configured to store at least (i) a semantic description relating to gaze-tracking data obtained during a user interaction involving one of the plurality of different computer applications and (ii) a semantic description relating to application event data generated as a result of another interaction involving another one of the plurality of different software applications and (iii) a recency or frequency of user attention to a user interface element, wherein the model is configured to initially store associations of semantic descriptions with the gaze-tracking data over time in a short-term memory portion of the model, and the model is configured to move at least a portion of the associations to a long-term memory portion of the model after a period of time; and an inference engine embodied in the one or more non-transitory machine-accessible media, the inference engine to; algorithmically draw an inference about a current user activity with the computing system based on the model; and make the inference available for use by a computer-executable module of the computing system. - View Dependent Claims (19, 20)
-
-
21. A system for modeling user attention to user interface elements displayed by a plurality of software applications on a computing device, the system embodied in one or more non-transitory machine-accessible storage media, the system comprising:
-
a contextual model comprising data relating to a plurality of user interactions involving a plurality of different software applications, the contextual model comprising semantic descriptions of each of the plurality of user interactions, each of the semantic descriptions relating to one or more of; a plurality of real-time sensor inputs received at the computing device, the real-time sensor inputs being indicative of a user'"'"'s gaze in relation to a display of the computing device; locations on the display corresponding to the real-time sensor inputs; the user interface elements displayed at the locations on the display corresponding to the real-time sensor inputs; and application event data;
wherein the contextual model is configured to initially store associations of the semantic descriptions with gaze-tracking data over time in a short-term memory portion of the contextual model, and the contextual model is configured to move at least a portion of the associations to a long-term memory portion of the model after a period of time; andan inference engine embodied in the one or more non-transitory machine-accessible media, the inference engine to; algorithmically draw an inference about a current user activity with the computing device based on (i) a semantic description relating to gaze-tracking data obtained during a user interaction involving one of the plurality of different software applications and (ii) another semantic description relating to application event data generated as a result of another interaction involving another one of the plurality of different software applications and (iii) a recency or frequency of user attention to a user interface element; and make the inference available for use by a computer-executable module of the computing device. - View Dependent Claims (22, 23, 24)
-
-
25. A method for modeling user activity with regard to a plurality of different software applications on a computing device, the method comprising, with the computing device, during a plurality of user interactions involving a plurality of different software applications over time:
-
receiving passive interaction data comprising one or more real-time sensor inputs indicative of a passive user interaction with the computing device, wherein the computing device is not configured to initiate any application events in response to the passive user interaction; receiving active interaction data indicative of an active user interaction, wherein the computing device is configured to initiate at least one application event in response to the active user interaction; generating a plurality of semantic descriptions relating to the passive interaction data and the active interaction data; developing a model of user activity in relation to the computing device based on associations of the plurality of semantic descriptions with gaze-tracking data over time, initially storing the associations in a short-term memory portion of the model, and moving at least a portion of the associations to a long-term memory portion of the model after a period of time; algorithmically drawing an inference about a current user activity with the computing device based on (i) a semantic description relating to gaze-tracking data obtained during a user interaction involving one of the plurality of different software applications and (ii) another semantic description relating to application event data generated as a result of another interaction involving another one of the plurality of different software applications and (iii) a recency or frequency of user attention to a user interface element; and making the inference available for use by a computer-executable module of the computing device.
-
-
26. A method for modeling user activity with regard to a plurality of different software applications on a computing device, the method comprising, with a computing system comprising one or more computing devices:
-
generating a semantic description for each of a plurality of user interactions with the computing device performed over time with a plurality of different types of software applications, said plurality of user interactions comprising at least one passive user interaction and at least one active user interaction, and wherein; for the at least one passive user interactions, said generating comprises; receiving gaze-tracking data comprising one or more real-time sensor inputs indicative of a location of a user'"'"'s gaze in relation to a display of the computing device; and generating a semantic description of the at least one passive user interaction based on information received from a one of the plurality of different software applications that is displaying content at the gaze location; and for the at least one active user interactions, said generating comprises; receiving, from at least one of the plurality of different software applications that is involved in said at least one active user interactions, a pre-defined textual description of an interactive function performed by the at least one of the plurality of software applications during the at least one active user interaction; and generating a semantic description of the at least one active user interaction based on the pre-defined textual description; developing a model of user activity in relation to the computing device based on associations of semantic descriptions with the gaze-tracking data over time, initially storing the associations in a short-term memory portion of the model, and moving at least a portion of the associations to a long-term memory portion of the model after a period of time; algorithmically drawing an inference about a current user activity with the computing device based on the semantic description of the at least one passive user interaction and the semantic description of the at least one active user interaction and a recency or frequency of user attention to a user interface element; and making the inference available for use by a computer-executable module of the computing system.
-
Specification