Automated disambiguation of users in a multi-tenant environment based on sensor data

A system, comprising:

• one or more processors;
• and
• memory storing thereon instructions that, as a result of being executed by the one or more processors, cause the system to:
• receive ambient and direct data corresponding to different sensors and devices within a multi-tenant environment, wherein the ambient and direct data is received through a gateway device implemented in the multi-tenant environment, and wherein the multi-tenant environment is associated with a set of users;
• identify one or more events occurring in the multi-tenant environment, wherein the one or more events are identified based on the ambient and direct data;
• extract attribute data from the ambient and direct data, wherein the attribute data corresponds to an individual detected within the multi-tenant environment, and wherein the individual is associated with the one or more events;
• dynamically process the attribute data through a trained classification algorithm to calculate an attribute vector, wherein attribute vectors are calculated to disambiguate the set of users according to a set of known attribute vectors associated with the set of users;
• calculate match probability scores corresponding to the set of users by determining differences between values of the attribute vector and known values corresponding to the set of known attribute vectors;
• identify a user from the set of users based on the match probability scores, wherein a match probability score associated with the user satisfies a score threshold value;
• associate the one or more events with the user;
• and
• dynamically update a known attribute vector associated with the user based on the one or more events and feedback corresponding to the one or more events, wherein the updated known attribute vector is used to better associate similar attribute data with the user.

A non-transitory computer-readable storage medium storing thereon executable instructions that, as a result of being executed by one or more processors of a computer system, cause the computer system to:

• receive ambient and direct data corresponding to different sensors and devices within a multi-tenant environment, wherein the ambient and direct data is received through a gateway device implemented in the multi-tenant environment, and wherein the multi-tenant environment is associated with a set of users;
• identify one or more events occurring in the multi-tenant environment, wherein the one or more events are identified based on the ambient and direct data;
• extract attribute data from the ambient and direct data, wherein the attribute data corresponds to an individual detected within the multi-tenant environment, and wherein the individual is associated with the one or more events;
• dynamically process the attribute data through a trained classification algorithm to calculate an attribute vector, wherein attribute vectors are calculated to disambiguate the set of users according to a set of known attribute vectors associated with the set of users;
• calculate match probability scores corresponding to the set of users by determining differences between values of the attribute vector and known values corresponding to the set of known attribute vectors;
• identify a user from the set of users based on the match probability scores, wherein a match probability score associated with the user satisfies a score threshold value;
• associate the one or more events with the user;
• and
• dynamically update a known attribute vector associated with the user based on the one or more events and feedback corresponding to the one or more events, wherein the updated known attribute vector is used to better associate similar attribute data with the user.