Design And Implementation Of Expert Recommendation System Based On Deep Learning And Community Question Answering

With the rapid development of Internet technology,the number of interactive posts on community forums has increased significantly.Community forums have become one of the main sources for people to obtain information.These Q&A Posts contain a lot of information exchange,helping users solve many problems.However,in some Q&A posts,there are some invalid or even wrong answers.Such answers may mislead users or people who pay close attention to the same questions,and even bring great losses to users.In order to efficiently solve the problems raised by users,,it is worthy to study how to find out relevant users to answer the question and reduce the impact of invalid answers on users when a new question is posted.Based on these questions,this paper presents following research contents.First of all,this paper proposes a correct answer recognition method based on community forum Q&A,combining both text features and non-text features,such as the number of words answered and the number of followers of the user who posted the answer.Through experiments,it has been proved that the method proposed in this paper has a good performance in the correct answer identification task.Secondly,in order to improve the accuracy of the expert recommendation function,this paper proposes a Q&A oriented keyword extraction algorithm which is inspired by the training model to solve the question that keyword extraction technology of supervision model generalization ability are relatively weak.This algorithm can effectively increase the robustness and accuracy of keyword extraction method.Finally,this paper designs an expert recommendation system based on deep learning and community forum Q&A.We analyzed the user needs of the system and distinguished the functional modules of the system.The visual interface of the system is also designed to improve user experience.