Back to results
US 11,599,890 B1Granted
Holistic fraud cocoon
AssigneeWells Fargo Bank, N.A.
InventorsChris Kalaboukis, Andrew J. Garner, IV, Andres J. Saenz, Abhijit Rao, Ramanathan Ramanathan
Life of this patent
Estimating the term…
Abstract
Provided herein is a holistic fraud cocoon that protects a user from incoming or outgoing fraud and phishing communications by capturing, analyzing and either altering, discarding or presenting communications to the user. For example, the holistic fraud cocoon captures all incoming sources (mail, in-person visits, phone, TV, radio, newspapers and internet) and the actions performed by the user, triggered by those sources. Certain actions are held in a queue to be reviewed, validated and acted upon by a third party before the actions are executed.
Claims18
A method, comprising:
- monitoring, by a system comprising a processor, an environment associated with an identified individual by using a set of sensors installed within the environment;
- capturing, based on the set of sensors monitoring the environment, filterable communication data associated with the identified individual, wherein the filterable communication data includes a first communication that includes audio data corresponding to at least part of a verbal communication being transmitted by a first another entity, and wherein:
- an audio sensor of the set of sensors captures the audio data of the first communication in response to detecting a trigger event, the trigger event including a detection of key words or phrases, voice inflection, and voice stress during the verbal communication;
- processing, by the system, the filterable communication data using a machine-learning model to generate a first classification label indicating an extent to which the filterable communication data corresponds to a predicted fraudulent activity;
- before the identified individual accesses the filterable communication data, performing, by the system, a first corrective action determining that the first classification label of the filterable communication data indicates the predicted fraudulent activity, where the first corrective action includes discarding the filterable communication data to prevent the identified individual from receiving the filterable communication data;
- accessing, by the system, a first feedback for the first corrective action;
- adjusting, by the system, one or more parameters of the machine-learning model based on the first feedback;
- capturing, based on the set of sensors monitoring the environment, unfilterable communication data associated with the identified individual, wherein the unfilterable communication data includes a second communication that includes text data of a physical document generated by a second entity, wherein the unfilterable communication data indicates that the identified individual has accessed the second communication, and wherein:
- a video sensor of the set of sensors captures the text data of the second communication in response to an interactive chatbot prompting the identified individual to hold one or more pages of the physical document in front of the video sensor;
- processing, by the system, the unfilterable communication data using the machine-learning model to generate a second classification label indicating an extent to which the unfilterable communication data corresponds to another predicted fraudulent activity;
- performing, by the system, second corrective actions based on determining that the second classification label of the unfilterable communication data indicates the other predicted fraudulent activity, wherein the second corrective action include:
- (i) intercepting and discarding one or more outgoing communications transmitted by the identified individual to prevent the second entity from accessing the one or more outgoing communications;
- and (ii) sending an alert to the identified individual to discard the physical document;
- identifying, by the system, a second feedback of the second corrective action;
- and
- further adjusting, by the system, the one or more parameters of the machine-learning model based on the second feedback.
13 more elements
A non-transitory computer-readable storage device that stores executable instructions that, in response to execution, cause a system comprising a processor to perform operations, comprising:
- monitoring an environment associated with an identified individual by using a set of sensors installed within the environment;
- capturing, based on the set of sensors monitoring the environment, filterable communication data associated with the identified individual, wherein the filterable communication data includes a first communication that includes audio data corresponding to at least part of a verbal communication being transmitted by a first entity, wherein the filterable communication data indicates that the identified individual is yet to access the first communication from the first entity, and wherein:
- an audio sensor of the set of sensors captures the audio data of the first communication in response to detecting a trigger event, the trigger event including a detection of key words or phrases, voice inflection, and voice stress during the verbal communication;
- processing the filterable communication data using a machine-learning model to generate a first classification label indicating an extent to which the filterable communication data corresponds to a predicted fraudulent activity;
- before the identified individual accesses the filterable communication data, performing a first corrective action based on determining that the first classification label of the filterable communication data indicates the predicted fraudulent activity, wherein the first corrective action includes discarding the filterable communication data to prevent the identified individual from receiving the filterable communication data;
- accessing a first feedback for the first corrective action;
- adjusting one or more parameters of the machine-learning model based on the first feedback;
- capturing, based on the set of sensors monitoring the environment, unfilterable communication data associated with the identified individual, wherein the unfilterable communication data includes a second communication that includes text data of a physical document generated by a second entity, wherein the unfilterable communication data indicates that the identified individual has accessed the second communication, and wherein:
- a video sensor of the set of sensors captures the text data of the second communication in response to an interactive chatbot prompting the identified individual to hold one or more pages of the physical document in front of the video sensor;
- processing the unfilterable communication data using the machine-learning model to generate a second classification label indicating an extent to which the unfilterable communication data corresponds another predicted fraudulent activity;
- performing second corrective actions based on determining that the second classification label of the unfilterable communication data indicates the other predicted fraudulent activity, wherein the second corrective actions include:
- (i) intercepting and discarding one or more outgoing communications transmitted by the identified individual to prevent the second entity from accessing the one or more outgoing communications;
- and (ii) sending an alert to the identified individual to discard the physical document;
- identifying a second feedback of the second corrective action;
- and
- further adjusting the one or more parameters of the machine-learning model based on the second feedback.
13 more elements
A system, comprising:
- a processor;
- and
- a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations, comprising:
- monitoring an environment associated with an identified individual by using a set of sensors installed within the environment;
- capturing, based on the set of sensors monitoring the environment, filterable communication data associated with the identified individual, wherein the filterable communication data includes a first communication that includes audio data corresponding to at least part of a verbal communication being transmitted by a first entity, wherein the filterable communication data indicates that the identified individual is yet to access the first communication from the first entity, and wherein:
- an audio sensor of the set of sensors captures the audio data of the first communication in response to detecting a trigger event, the trigger event including a detection of key words or phrases, voice inflection, and voice stress during the verbal communication;
- processing the filterable communication data using a machine-learning model to generate a first classification label indicating an extent to which the filterable communication data corresponds to a predicted fraudulent activity;
- before the identified individual accesses the filterable communication data, performing a first corrective action based on determining that the first classification label of the filterable communication data indicates the predicted fraudulent activity, wherein the first corrective action includes discarding the filterable communication data to prevent the identified individual from receiving the filterable communication data;
- accessing a first feedback for the first corrective action;
- adjusting one or more parameters of the machine-learning model based on the first feedback;
- capturing, based on the set of sensors monitoring the environment, unfilterable communication data associated with the identified individual, wherein the unfilterable communication data includes a second communication that includes text data of a physical document generated by a second entity, wherein the unfilterable communication data indicates that the identified individual has accessed the second communication, and wherein:
- a video sensor of the set of sensors captures the text data of the second communication in response to an interactive chatbot prompting the identified individual to hold one or more pages of the physical document in front of the video sensor;
- processing the unfilterable communication data using the machine-learning model to generate a second classification label indicating an extent to which the unfilterable communication data corresponds another predicted fraudulent activity;
- performing second corrective actions based on determining that the second classification label of the unfilterable communication data indicates the other predicted fraudulent activity, wherein the second corrective actions include:
- (i) intercepting and discarding one or more outgoing communications transmitted by the identified individual to prevent the second entity from accessing the one or more outgoing communications;
- and (ii) sending an alert to the identified individual to discard the physical document;
- identifying a second feedback of the second corrective action;
- and
- further adjusting the one or more parameters of the machine-learning model based on the second feedback.
16 more elements
Dependencies are inferred from the claim text.