User profile based on clustering tiered descriptors

A non-transitory machine-readable storage medium comprising instructions that, when executed, cause at least one processor to at least:

• group descriptors into a cluster of descriptors based on an association between the descriptors and each of a first item and a second item;
• access, via a user device communicatively coupled to the machine via a network, biometric data of a user;
• determine a first activity in which the user is engaged based on contextual data that correlates the cluster of descriptors with the biometric data of the user received from the user device via the network;
• determine whether the first activity represents an anomalous phase of the user based on a time period of the first activity having a duration shorter than a threshold duration;
• in response to determining that the first activity does not represent the anomalous phase of the user, generate a user profile based on the first activity of the user and the cluster of descriptors;
• in response to determining that the first activity represents the anomalous phase of the user, reduce, by executing an instruction with a processor, the size of user profile by omitting a name of the cluster of descriptors from the user profile;
• and
• when a second activity of the user matches the first activity associated with the cluster of descriptors within the user profile, generate a recommendation including a third item based on the user profile.

A method comprising:

• grouping, by executing an instruction with a processor, descriptors into a cluster of descriptors based on an association between the descriptors and each of a first item and a second item;
• accessing, via a user device communicatively coupled to the processor via a network, biometric data of a user;
• determining, by executing an instruction the processor, a first activity in which the user is engaged based on contextual data that correlates the cluster of descriptors with the biometric data of the user received from the user device via the network, determining whether the first activity represents an anomalous phase of the user based on a time period of the first activity having a duration shorter than a threshold duration;
• in response to determining that the first activity does not represent the anomalous phase of the user, generating, by executing an instruction the processor, a user profile based on the first activity of the user and the cluster of descriptors;
• in response to determining that the first activity represents the anomalous phase of the user, reducing, by executing an instruction with a processor, the size of user profile by omitting a name of the cluster of descriptors from the user profile;
• and
• generating, by executing an instruction the processor, in response to a second activity of the user matching the first activity associated with the cluster of descriptors within the user profile, a recommendation including a third item based on the user profile.