Question recommending method, apparatus and system

A question recommending apparatus, comprising:

• an interest vector calculation module, configured to construct an interest vector of a user according to a short-term interest, a long-term interest, and a group interest of the user, the interest vector of the user comprising multiple interest items, wherein the interest vector calculation module further comprises:
• an interest vector calculation sub-unit, configured to combine the short-term interest, the long-term interest, and the group interest of the user according to a vector weighting coefficient of the short-term interest of the user, a vector weighting coefficient of the long-term interest of the user, and a vector weighting coefficient of the group interest of the user and by means of linear weighting of vectors, to obtain the interest vector of the user;
• a candidate-for-recommendation question acquiring module, configured to acquire at least one candidate-for-recommendation question from a set of to-be-resolved questions according to the multiple interest items;
• a to-be-recommended question acquiring module, configured to separately estimate, according to a click-through rate (CTR) model, a probability of answering the candidate-for-recommendation question by the user, and select a to-be-recommended question from the at least one candidate-for-recommendation question according to the estimated probability;
• and
• a question recommending module, configured to recommend the to-be-recommended question to a client of the user.