Typifying emotional indicators for digital messaging
AssigneeCapital One Services, LLC
InventorsJeremy Phillips, Andrew Beane
Life of this patent
Estimating the term…
Abstract
The present disclosure provides computing systems and techniques for indicating an emotional and/or environmental state of a user in a digital messaging application. A computing device can determine an emotional and/or environmental state of a first user responsive to reading or responding to a message and can convey the determined emotional and/or environmental state to a second computing device, to be transiently presented by the second computing device.
Claims20
An apparatus, comprising:
- a processor;
- and
- a memory coupled to the processor, the memory comprising instructions that when executed by the processor cause the processor to:
- determine state data based on characteristics of interaction by a user with a messaging device and characteristics of the messaging device in response to the interaction;
- generate a state indication based in part on the state data via a machine learning model, wherein the state indication comprises an indication of an emotional state of the user of the messaging device, an indication of an environmental state of the user of the messaging device, or both the indication of the emotional state and the indication of the environmental state;
- provide the state indication to a digital message application of the message device;
- receive a state indication feedback associated with the state indication;
- and
- train the machine learning model with the state indication feedback to create a trained state prediction model to predict state indications.
A non-transitory computer readable medium embodied with programming code executable by a processor that cause the processor to:
- determine state data based on characteristics of interaction by a user with a messaging device and characteristics of the messaging device in response to the interaction;
- generate a state indication based in part on the state data via a machine learning model, wherein the state indication comprises an indication of an emotional state of the user of the messaging device, an indication of an environmental state of the user of the messaging device, or both the indication of the emotional state and the indication of the environmental state;
- provide the state indication to a digital message application of the message device;
- receive a state indication feedback associated with the state indication;
- and
- train the machine learning model with the state indication feedback to create a trained state prediction model to predict state indications.
A method comprising:
- determining state data based on characteristics of interaction by a user with a messaging device and characteristics of the messaging device in response to the interaction;
- generating a state indication based in part on the state data via a machine learning model, wherein the state indication comprises an indication of an emotional state of the user of the messaging device, an indication of an environmental state of the user of the messaging device, or both the indication of the emotional state and the indication of the environmental state;
- providing the state indication to a digital message application of the message device;
- receiving a state indication feedback associated with the state indication;
- and
- training the machine learning model with the state indication feedback to create a trained state prediction model to predict state indications.
Dependencies are inferred from the claim text.