Personalization of recommendations based on building model and behavioral science

A system for providing personalized building recommendations, the system comprising:

• one or more processors;
• a memory coupled to the one or more processors;
• a profile builder configured to:
• receive a first set of data elements representing features of a building;
• receive a second set of data elements representing user characteristics;
• create, using the first set of data elements, a building profile comprising:
• a set of physical properties of the building;
• and
• a building model that, when operated under various condition parameters, represents how buildings with the set of physical properties will act at different times;
• and
• create, using the second set of data elements, a behavioral model that comprises building user details and action predictors, the action predictors comprising determined user characteristics, history of user actions, or probabilities of user actions;
• a recommendation personalization engine configured to:
• select, based on the building profile and the behavioral model, one or more recommendation filters wherein each recommendation filter is operable to select, remove, or modify one or more elements from a set of possible recommendations;
• choose, by operating the selected recommendation filters, a set of filtered recommendations;
• create building scenarios by—
• calculating, for one or more filtered recommendation subsets of the set of filtered recommendations, condition parameters corresponding to the filtered recommendation subset;
• and
• operating the building model under the calculated condition parameters corresponding to the recommendation subsets;
• wherein each building scenario corresponds to a filtered recommendation subset;
• and
• select probable recommendations based on determinations from the behavioral model, operating on the set of filtered recommendations, the determinations indicating that one or more filtered recommendations will be helpful, followed, or found valuable to the user;
• and
• an output generator configured to:
• select one or more elements from the probable recommendations, building scenarios, and set of filtered recommendations;
• and
• create and transmit, from the selected elements, output elements.

A computer-readable storage medium storing instructions that, when executed by a computing device, cause the computing device to perform operations for providing personalized building recommendations, the operations comprising:

• receiving a first set of data elements representing features of a building;
• receiving a second set of data elements representing user characteristics;
• creating, using the first set of data elements, a building profile comprising:
• a set of physical properties of the building;
• and
• a building model that, when operated under various condition parameters, represents how buildings with the set of physical properties will act at different times;
• creating, using the second set of data elements, a behavioural model that comprises building user details and action predictors, the action predictors comprising determined user characterises, history of user actions, or probabilities of user actions;
• selecting, based on the building profile and the behavioural model, one or more recommendation filters wherein each recommendation filter is operable to select, modify, or remove one or more elements from a set of possible recommendations;
• choosing, by operating the selected recommendation filters, a set of filtered recommendations;
• creating building scenarios by—
• calculating, for one or more filtered recommendation subsets of the set of filtered recommendations, condition parameters corresponding to the filtered recommendation subset;
• and
• operating the building model under the calculated condition parameters corresponding to the recommendation subsets;
• wherein each building scenario corresponds to a filtered recommendation subset;
• selecting probable recommendations based on determinations from the behavioral model, operating on the set of filtered recommendations, the determinations indicating that one or more filtered recommendations will be helpful, followed, or found valuable to the user;
• selecting one or more elements from the probable recommendations, building scenarios, and set of filtered recommendations;
• and
• creating and transmitting, from the selected elements, output elements.