Method and system for recommending content items to a user based on tensor factorization

A system having at least one processor, storage, and a communication platform connected to a network for recommending content items to a user, comprising:

• a recommendation request analyzer configured for receiving a request for recommending content items to the user;
• a user tensor data retriever configured for obtaining tensor data related to a plurality of users and a plurality of content items based on the request;
• a model bound generator configured for determining at least one bound for a tensor factorization model that is generated based on a prior probability distribution;
• a distributed inference engine configured for decomposing the tensor data into a plurality of sub-tensors based on the prior probability distribution and predicting one or more items interesting to the user based on the at least one bound and the plurality of sub-tensors;
• and
• a content recommendation engine configured for recommending at least one of the one or more items to the user as a response to the request.

A non-transitory machine-readable medium having information recorded thereon for recommending content items to a user, wherein the information, when read by the machine, causes the machine to perform the following:

• receiving a request for recommending content items to the user;
• obtaining tensor data related to a plurality of users and a plurality of content items based on the request;
• decomposing the tensor data into a plurality of sub-tensors based on a prior probability distribution;
• determining at least one bound for a tensor factorization model that is generated based on the prior probability distribution;
• predicting one or more items interesting to the user based on the at least one bound and the plurality of sub-tensors;
• and
• recommending at least one of the one or more items to the user as a response to the request.