Research And Implementation Of Intelligent Question Answering Model For College Admission Counseling

Intelligent question answering system combines natural language processing technology and information retrieval technology,and can answer users’ questions in natural language accurately and concisely.According to its application field,the question answering system can be divided into limited domain question answering system and open domain question answering system.In this thesis,a question answering system model in a limited field is designed,which focuses on studying and improving the question and answer matching model,fully understands the Chinese semantics,and proposes the question and answer matching model of Si Bi LA(Siamese Bidirectional Long Short Term Memory Attention)to improve the accuracy of the question answering system.In order to improve the time efficiency of the question answering system,doc2 vec sorting algorithm is finally introduced into Si Bi LA.This question answering model can be applied to the field of college admission counseling,so as to save the human resources and cost of college admission counseling.The main work of this thesis is as follows:Firstly,a question and answer matching model Siamese Bidirectional Long Short Term Memory model based on Bidirectional LSTM and siamese neural network is proposed.This model uses bidirectional LSTM to capture the bidirectional semantic dependency of statements and introduces the siamese neural network structure.Si Bi L model can make full use of the semantic information and word order information between words in the text,and deal with the similarity,polysemy and order of Chinese words,thus improving the accuracy and time efficiency of the question answering system.Secondly,in order to further improve the accuracy of the question answering system,attention mechanism was added to the Si Bi L model to build a new question and answer matching model,Si Bi LA.The attention mechanism can assign different weights to the word vectors according to the importance of the words in the sentence,which can better highlight the representational ability of the central word in the sentence,so as to further improve the accuracy of the question answering model.In order to further reduce the number of model matching,this paper uses the sorting algorithm based on doc2 vec before the question and answer matching model,so as to improve the time efficiency of the question answering systemFinally,the intelligent question answering system for college admission counseling based on Si Bi LA question answering model is established,the demand analysis of the intelligent question answering system for college admission counseling is completed,the architecture design and the design of main functional modules are summarized,and the intelligent question answering system for college admission counseling is realized by combining with the database of college admission counseling questions.