Intelligent Algorithm Analysis And Design Of Video Recommendation System

With the rapid development of Internet information technology throughout the world,humanity has gradually entered the era of information overload,resulting in the Internet information is difficult to find by interested users in a timely manner,and consumers of information are also difficult to access from numerous data.Find what you really like.Therefore,the recommendation of network information for user interest has become one of the hot topics in the current research.With the development of network video technology and the continuous enrichment of video information,how to recommend interesting video content for users from the vast video is the focus of this article.This article recommends interesting videos for users based on user information,video information,and user rating information for videos.It focuses on the analysis of the advantages and disadvantages of video recommendation intelligent algorithms based on BP neural networks,and proposes BP neural networks optimized based on bee colony intelligence algorithms.The network and the intelligent video recommendation algorithm based on reinforcement learning BP neural network add these intelligent algorithms to the video recommendation system to achieve the goal of improving the recommendation effect of the video recommendation system.The main work of this article includes:Aiming at the problem of large data and complex sample space in the current video recommendation system,principal component analysis(PCA)is used to reduce the dimension of the sample space in the video recommendation system by analyzing the correlation between data attributes in the sample.To represent the original data,to achieve the goal of reducing the number of neurons in the input layer of the BP neural network,and to reduce the complexity of the video recommendation algorithm.Aiming at the problem that the video recommendation algorithm based on BP neural network is easy to fall into local minimum value and the convergence speed is slow,the BP neural network video recommendation algorithm optimized by bee colony intelligence algorithm is proposed,and the weight and threshold of neural network based on bee colony intelligence optimization algorithm are used.Optimizing to avoid the random selection of BP neural network on the network weights and thresholds leads to the problem that the recommendation system falls into a localoptimum and the convergence speed is slow,and a globally optimal video recommendation system is realized.Aiming at the problem that the structure of network hidden layer nodes in BP neural network video recommendation algorithm is difficult to determine,a BP neural network video recommendation algorithm based on reinforcement learning intelligence algorithm is proposed.By strengthening the learning intelligence algorithm to adjust the structure of neural network hidden layer,add and delete hidden layers.The number of nodes achieves the goal of an optimized network hidden layer structure and achieves the goal of improving the accuracy of the video recommendation algorithm.In order to verify the effectiveness of BP neural network video recommendation based on intelligent algorithm optimization,this paper uses Movielens data set to verify the algorithm,and uses the accuracy,mean square error,confusion matrix,cross-entropy loss function and other parameters to evaluate the experimental results.,to achieve the goal of improving video recommendations.