| The explosive growth of Internet information has not only provided users with rich information and knowledge,but also caused great difficulties for users to screen the knowledge they need.How to return correct information for users among a large amount of information in various structures has become a hot topic in current knowledge retrieval.Therefore,this article aims to develop a question and answer system for the medical and health field on the YSU-II rehabilitation robot studied in the laboratory.The main work of this article is as follows.First of all,in order to make this system more intelligent and reduce the barriers for users to use this system,by calling the HKUST XunFei speech recognition interface and speech synthesis interface,respectively,online audio recognition and online conversion are realized to complete the information recognition and broadcast functions,thus Realize the process of human-computer interaction.Secondly,for the extraction of disease symptom entities,an improved similarity calculation method based on edit distance and a similarity calculation method based on word vectors are combined.This method is combined with AC multi-pattern matching algorithm to complete The task of entity extraction of disease symptom inthis system.Then,for the practical application of the YSU-II rehabilitation robot,multiple data sources are used to implement the medical question and answer system,in which the graph database is used to complete the relevant medical question and answer,and the relational database is used to implement the question and answer of the relevant body parts.Finally,the entire medical assistant question and answer system was transplanted to the YSU-II rehabilitation robot using Labview graphical editing language.Using the medical assistant question and answer system in this article to search and query some diseases,body joints and other issues,give users some health advice or help users always understand their own physical conditions,etc.,played a certain auxiliary role.This paper implements the construction and performance evaluation of the medical assistant question and answer system.The performance evaluation shows that the recall rate and accuracy of the system are high,and a satisfactory question and answer effect is achieved. |