Design And Implementtation Of Lawyer Recommendation System Based On Deep Learning

With the construction and advancement of China’s legal society,legal awareness has been deeply rooted in the people’s mind,and it has become common sense to protect the legitimate and legitimate interests of individuals through legal means.However,due to the strong legal professionalism,individuals need to seek the help of lawyers to make professional defense when solving problems through legal channels,so it is very meaningful to provide lawyer recommendations for users.At the same time,because the judgment documents have been published legally on the Chinese judgment documents,it provides a lot of text data for the realization of the lawyer recommendation system.On the basis of deep learning and recommendation algorithm research,this paper combines the two to build the recommendation algorithm based on deep learning used in this paper.The design of the system is completed by taking a large number of open judgment documents as rich text data,providing the required data for the recommendation algorithm through natural language processing technology.Firstly,the function of classifying judge documents is designed,and the judge documents are classified by multi-label algorithm.Then designed to extract the lawyer information and decision functions,will be divided after the verdict after according to the article,the design presented in this paper the BiGRU the verdict orientation analysis algorithm based on attention mechanism to deal with the verdict,analyze the verdict of the tendency,and the results for the grading of lawyers,combination of extracting information construct lawyer library.By directly or indirectly obtaining user needs,the convolutional neural network in deep learning is used to extract the characteristics of user needs and lawyer information.Finally,after similarity calculation of the two,the recommendation list is selected to realize the task of recommending lawyers for users.Through the functional test,performance test and system evaluation of the system,it is confirmed that the system meets the needs of users,runs stably and achieves the expected goals.