Research On Text Retrieval And Recommendation System Based On Deep Learning

With the rapid development of the Internet,which has led to a sharp increase in Internet news.How to accurately,quickly and effectively obtain the news that users are interested in from the mass Internet media has become an urgent problem to be sloved.Traditional text retrieval algorithms only calculate the relevance of search terms and texts,and obtain retrieval results according to the ranking of scores.The problem is lack of interaction with the user's historical behavior.In addition,traditional recommendation algorithms have shortcomings such as excessive manual intervention and difficulty in extracting feature information.Therefore,in order to solve above problems.This thesis studies the application of deep learning in retrieval and recommendation algorithms.The main work of this thesis includes the following aspects:1.On the basis of analyzing and researching existing retrieval algorithms,a retrieval algorithm based on historical interaction behavior of users is designed.This thesis adopts a text retrieval method that combines traditional retrieval algorithms and unsupervised clustering algorithms,designs a user interactive retrieval model,optimizes the model parameters,and realizes the retrieval of news information.This thesis conducts a comparative experiment on the algorithms and studies the influence of parameters on the retrieval results.The results show that the new algorithm improves the accuracy of retrieval news.2.The recommendation algorithm based on deep learning is studied.Based on the news feature extraction algorithm combines extraction and generation,a news recommendation model based on the Bert model is designed.And it achieved the accurate recommendation of the Internet news.The experiments to verify the effectiveness of the recommendations is designed.3.On the basis of researching news retrieval and recommendation algorithms fused with deep learning,a news recommendation retrieval and recommendation system based on Internet media is designed and implemented.The system mainly includes: text management,model training,news retrieval,news recommendation,user management and other core functional modules.At the same time,the system was deployed and tested.