An Algorithm of Collaborative Recommendation Based on User's Interest Sub-Class

With the development of E-commerce, the magnitudes of users and Web documents grow rapidly, and result in the extreme data sparseness of users. The traditional algorithms of collaborative recommendation can't solve the problem very well. To address this issue, a novel algorithm of collaborative recommendation based on users' interest sub-classes is proposed. Based on the similarity of the interest sub-classes among the users, the new method makes it more easy and accurate to find the similar neighbors of a user, even if their interests are very different as a whole, and can provide more efficient information recommendation. Our experiment shows that this method can efficiently solve the problem of the extreme data sparseness of users, and can provide better result on information recommendation than the traditional algorithm of collaborative recommendation.