Methods and Apparatuses for Predicting User Destinations

Apparatus for predicting a destination for a user, the apparatus comprising circuitry configured to

• determine, based on historic search data associated with the user, at least one candidate destination based on the user and a given context;
• determine a plurality of embedding vectors associated with the at least one candidate destination, the user, and the given context from an embedding matrix, the embedding matrix comprising embedding vectors for different components of the historic search data;
• feed the plurality of embedding vectors into one or more first neural network layers configured to generate a semantic embedding for the candidate destination;
• and
• feed the semantic embedding into one or more second neural network layers configured to generate a probability score for the candidate destination.

A vehicle, comprising

• circuitry configured to
• determine, based on historic search data associated with the user, at least one candidate destination based on the user and a given context;
• determine a plurality of embedding vectors associated with the at least one candidate destination, the user, and the given context from an embedding matrix, the embedding matrix comprising embedding vectors for different components of the historic search data;
• feed the plurality of embedding vectors into one or more first neural network layers configured to generate a semantic embedding for the candidate destination;
• and
• feed the semantic embedding into one or more second neural network layers configured to generate a probability score for the candidate destination,
• wherein the circuitry is configured to
• generate a respective probability score for a plurality of candidate destinations and to predict a user's target destination by choosing the candidate destination having the highest probability score.