Design And Implementation Of Automatic Question Answering System

In recent years,the rise in popularity of the web has created a demand for services which help users to skip over all irrelevant information quickly.One of the services is question answering(QA),the technique of providing answers to specific questions.During the last decade,a number of automatic QA systems have emerged,which has been largely driven by the TREC(Text Retrieval Conference)QA Track.The technology of QA relates to a lot of aspects of NLP(Natural Language Processing),such as Information Retrieval(IR),Information Extraction(IE),Automatic Summarization,Conversation Interface,etc.However,recently the QA systems have emerged following two directions:open domain QA system and restricted domain QA system.Open domain QA systems deal with questions about nearly everything and can only rely on general ontology and world knowledge.Alternatively,unlimited types of questions are accepted in open domain question answering system.Restricted domain QA systems deal with questions under a specific domain(for example,biomedicine or weather forecasting)and can be seen as an easier task because NLP systems can exploit domain-specific knowledge frequently formalized in ontology.Alternatively,limited types of questions(questions related to q particular domain)are accepted in restricted domain.In this thesis,the aim is to propose SieQA,a restricted domain QA System addressed to School of International Exchange of Northeastern University,which combines sentence analysis,sentence classification and similarity computation to retrieve answer in our database(consists of the pairs of questions and answers).This thesis describes the general architecture of a QA system and discusses about the different steps in retrieving the answer to a question.Based on the method of automated question answering,and according to the business requirement of the School of International Exchange of Northeastern University,the thesis describes the system analysis which includes functional and nonfunctional requirement analysis.Then,the system architecture design,logical design,key modules design and database design are presented in this thesis.The SieQA was developed following a 3-tier architecture which has the client layer,the web App layer and the data layer.The client layer is an android chat application.It offers the user,a graphical interface allowing to log into the system,to ask questions,receive answers and browse the history of his asked questions.The web application was developed with tools like Wicket,spring,hibernate,JPA.It constitute the administration part of the system.It provides different web pages that allow the logged in user to manage domains and subdomains questions,questions,users and their roles and respond to unanswered questions.It offers a global view of the system statistics.The web application also integrates a component that handles users' requests,respond automatically to the asked questions and send notifications to the system administration in a failure case.This thesis also carries out several experiments of answer retrieval,whose results,shows high accuracy and proves that SieQA can help Northeastern University foreign students find answers to their questions.