Hybrid Recommendation Algorithm Combining Content and AHP
Guang Wang, Wang Zhen-bo · 2023
Personalized recommendation system is an effective way to help users quickly obtain demand data in the era of information overload, but there are still problems such as sparse data, cold start, inaccurate similarity calculation, etc. In order to solve the above problems, this paper proposes a hybrid recommendation algorithm that combines content and analytic hierarchy process. First, the user profile constructed by user information, project information and scoring information is fused with the content based recommendation algorithm. Then, we use AHP to improve the similarity calculation and prediction score of the traditional collaborative filtering algorithm. Finally, the content recommendation algorithm based on user profile and the improved collaborative filtering algorithm are mixed in a weighted way to obtain a hybrid recommendation algorithm that combines content and analytic hierarchy process. Experiments show that, compared with other comparison algorithms, the algorithm proposed in this paper can overcome the limitations of a single algorithm, effectively improve the accuracy of the recommendation results, and thus improve the recommendation quality.