Research And Implementation Of Intelligent Question And Answer System Based On Knowledge Graph In Cultural Field

Knowledge graph organizes the knowledge of the objective world in a structured form.They are rich in semantic information and are applied in more and more vertical fields.In the professional field,knowledge maps can be combined with question-and-answer systems,which can not only improve the acquisition efficiency of professional knowledge,serve domain experts,but also reduce the threshold of professional knowledge and face the general public.This article focuses on the field of traditional culture,and studies the construction of knowledge maps in the cultural field and the implementation of intelligent question-and-answer systems.The main work completed includes:1.Research on core ontology and knowledge graph construction in the cultural field.Based on the application purpose of intelligent question and answer,comprehensively considering the professionalism and public understanding of knowledge,the concept of universal core ontology in the cultural field related to the application of this article is analyzed.Taking Beijing Opera ontology as an example,the expansion of the core ontology is tested,and the construction of knowledge graph in cultural sub-fields such as Beijing Opera,architecture,and cultural relics is completed through data collection,knowledge extraction and knowledge storage.2.Design and implementation of intelligent question answering algorithm based on knowledge graph in the cultural field.The algorithm takes named entity recognition and relational attribute mapping as the main part,and combines entity links to improve the practical application effect of the algorithm.For named entity recognition,this paper combines the BERT pre-training model of full-word Mask,and proposes a BERT-BiLSTM-CRF model for extracting question entities;for relational attribute mapping,this paper uses the BERT twin neural network model to perform the semantic similarity calculation of user questions and relational attributes,which is used to identify the user’s question intention.Finally,a query statement is constructed for knowledge graph retrieval to obtain answers.3.Design and implementation of interactive question-and-answer system based on knowledge graph in the cultural field.The system can receive natural language questions in the cultural field raised by users,return the corresponding answers after algorithm analysis,and present the knowledge network related to question-and-answer in a visual form,which plays a supporting role in question-and-answer results and knowledge recommendation.This paper focuses on the knowledge graph and intelligent question-and-answer in the cultural field,and the final question-and-answer system has a good application effect on the knowledge question-and-answer in the cultural field,which can help users query relevant knowledge in the cultural field.At the same time,the research of this paper is also an effective exploration of the form of cultural digital services.