【摘要】 提高推荐系统的推荐性能过去一直是一个非常大的挑战,因为在进行推荐的时候要同时兼顾推荐结果的准确性和计算推荐结果的计算时间。基于上述问题,提出一种基于深度学习的推荐算法,通过深度学习的方法挖掘用户和电影的特征、训练模型,从而提高推荐算法准确性。同时,通过神经网络提取用户和电影的特征,而不是基于用户对电影的评分矩阵,解决了推荐系统中的稀疏性问题和冷启动问题。最后在真实的数据集上进行实验,验证推荐算法的准确性。更多还原

【Abstract】 Improving the recommendation performance of the recommendation system has always been a very big challenge,because the accuracy of the recommendation results and the calculation time for calculating the recommendation results are taken into consideration when making the recommendation. Based on the above problems,we proposed a recommendation algorithm based on deep learning,which used the deep learning method to mine the characteristics of users,movies and trains the model to improve the accuracy of the recommendation algorithm. At the same time,the characteristics of users and movies were extracted through the neural network,instead of solving the sparsity problem and the cold start problem in the recommendation system based on the user’s scoring matrix for the movie. And an experiment on real data was set to verify the accuracy of the recommended algorithm.更多还原