Abstract: Methods and servers for training a Machine Learning Algorithm (MLA) to generate predicted collaborative embeddings are disclosed. The method includes generating a training set for a training item which includes (i) a target collaborative embedding generated by an other MLA based on previous user-item interactions that are sufficient for generating the target collaborative embedding, (ii) the training item. During a given training iteration of the MLA, the server (i) inputs the training item into the MLA that generates a predicted collaborative embedding, (ii) determines a penalty score by comparing the predicted collaborative embedding generated by the MLA and the target collaborative embedding generated by the other MLA, and (iii) adjusts the MLA using the penalty score so as to increase the similarity between the predicted collaborative embedding and the target collaborative embedding of the training item.

Abstract: A method and system for determining user-specific proportions of types of content for recommendation to a given user comprising: acquiring for each respective type of content of at least two types of content, a respective base interval of proportion of content for recommendation, computing for each respective type of content of the at least two types of content an associated respective distribution of user interaction parameters associated with a respective set of users, acquiring an associated respective user-specific interaction parameter of the given user, computing for each respective type of content a respective user-specific proportion of the respective type of content for content recommendation to the given user, the respective user-specific proportion being within the respective base interval of proportion of content, the computing being based on: the respective distribution of user interaction parameters of the set of users, and the respective user-specific interaction parameter of the given user.

Abstract: A method and system for determining user-specific proportions of types of content for recommendation to a given user comprising: acquiring for each respective type of content of at least two types of content, a respective base interval of proportion of content for recommendation, computing for each respective type of content of the at least two types of content an associated respective distribution of user interaction parameters associated with a respective set of users, acquiring an associated respective user-specific interaction parameter of the given user, computing for each respective type of content a respective user-specific proportion of the respective type of content for content recommendation to the given user, the respective user-specific proportion being within the respective base interval of proportion of content, the computing being based on: the respective distribution of user interaction parameters of the set of users, and the respective user-specific interaction parameter of the given user.

Abstract: A method and system for determining user-specific proportions of types of content for recommendation to a given user comprising: acquiring for each respective type of content of at least two types of content, a respective base interval of proportion of content for recommendation, computing for each respective type of content of the at least two types of content an associated respective distribution of user interaction parameters associated with a respective set of users, acquiring an associated respective user-specific interaction parameter of the given user, computing for each respective type of content a respective user-specific proportion of the respective type of content for content recommendation to the given user, the respective user-specific proportion being within the respective base interval of proportion of content, the computing being based on: the respective distribution of user interaction parameters of the set of users, and the respective user-specific interaction parameter of the given user.