Methods for predicting destinations from partial trajectories employing open-and closed-world modeling methods

A system that facilitates determining one or more destinations of a user on a trip, comprising:

• at least one computer storage medium comprising computer executable instructions that, when executed by a processor:
• receive input data associated with at least one of one or more priors and one or more likelihoods;
• and
• predict one or more destinations based on a combination of at least one of the one or more priors and at least one of the one or more likelihoods, each of the one or more priors indicating a probability that a location is a destination of the trip based on prior activity by the user in visiting locations, and each of the one or more likelihoods indicating a probability that a location is the destination of the trip based on trip data, each of the one or more likelihoods being a trip time likelihood based at least in part on an estimated time to the location and an elapsed trip time.

A system that facilitates determining one or more destinations of a user, comprising:

• at least one processor;
• an interface component comprising computer-executable instructions that, when executed by a processor of the at least one processor, receives input data from at least one sensor;
• and
• a destination estimator component comprising computer-executable instructions that, when executed by a processor of the at least one processor, probabilistically predicts one or more candidate destinations is a destination for a trip based on at least one prior and at least one likelihood based at least in part on the received input data, wherein:
• the received input data comprises an indication of an amount of time that a user has been traveling for a trip;
• the likelihood is an efficient driving likelihood, the efficient driving likelihood being related to a computed driving efficiency associated with the set of candidate destinations, the computed driving efficiency being based on a ratio between how much time the user should have spent moving toward the candidate destination, if the user were following an efficient route to the candidate destination, and how much time that the user has actually been traveling for the trip;
• and
• each of the one or more priors indicates a probability that a location is a destination of the trip based on activity by the user in visiting prior locations.