Methods and machine-learning systems that provide a personalized concierge service

An automated concierge service comprising:

• a cloud service, implemented by multiple application instances executing on multiple servers within a cloud-computing facility, each including one or more processors and one or more memories, the cloud service
• receiving requests from client-side applications running on user devices, such as user smart phones, and responding to the received requests,
• requesting and receiving information from user devices, web sites, information services, user computers, vendor computers, and other information sources,
• storing information in a database that includes representations of entities, including events, venues, user-consumable services, and items, the representations including attribute values for attributes that characterize the entities, and
• storing information in a database that includes representations of users, including user attribute values and weights associated with potential matches between user attributes and entity attributes, and
• employing machine learning to continuously adjust and optimize the weights, attribute values, and other information and logic used by the cloud service to provide recommendations to users;
• and
• client-side applications that run on users' device to provide an automated-concierge-service user interface through which users
• request recommendations,
• receive and view recommendations of events, venues, user-consumable services, and items,
• select events, venues, user-consumable services, and items from among the displayed recommendations of events, venues, user-consumable services, and items, and
• arrange to attend, participate in, and/or consume selected events, venues, user-consumable services, and items.
• a set of computational entities that include one or more processors, one or more memories, and one or more data-storage devices;
• and
• computer instructions that execute on one or more of the one or more processors and that implement
• a first component that
• receives data streams containing inventory data for each of multiple management services and/or applications that provide management functionalities with respect to a distributed-computing facility, and
• processes the received data streams to generate and maintain stored data that represents correlations between synthetic keys generated and used by the multiple management services and/or applications to identify computational resources in the distributed-computing facility, and
• a second component that
• receives requests from requesting computational entities,
• uses the stored data that represents correlations between the synthetic keys used by the multiple management services and/or applications to generate responses to the requests, and
• transmits the responses to the requesting computational entities.