Collaborative Filtering (CF) is widely used in large-scale recommendationengines because of its efficiency, accuracy and scalability. However, inpractice, the fact that recommendation engines based on CF require interactionsbetween users and items before making recommendations, make it inappropriatefor new items which haven't been exposed to the end users to interact with.This is known as the cold-start problem. In this paper we introduce a novelapproach which employs deep learning to tackle this problem in any CF basedrecommendation engine. One of the most important features of the proposedtechnique is the fact that it can be applied on top of any existing CF basedrecommendation engine without changing the CF core. We successfully appliedthis technique to overcome the item cold-start problem in Careerbuilder's CFbased recommendation engine. Our experiments show that the proposed techniqueis very efficient to resolve the cold-start problem while maintaining highaccuracy of the CF recommendations.
展开▼
