Research And Application Of Construction Method Of Historical Cultural Relic Knowledge Map Based On Deep Learning

With the rapid development of internet technology,how to efficiently organize,manage and use digital resources of cultural relics has become a focus of attention in recent years.During the exhibition process of domestic museums,it is difficult to achieve effective organization,management,and management of cultural relics data through traditional methods.Some of these museums also manage cultural relics data by temporarily constructing small-scale knowledge maps.These small-scale knowledge maps of historical relics need to be manually constructed manually and lack intelligent methods.Based on the above background,from the actual point of view,this paper will introduce the relevant methods of deep learning to construct the knowledge map of historical cultural relics,and conduct in-depth research and analysis on the entity recognition model and algorithm in the construction of the knowledge map of cultural relics.Improves the accuracy of model checking.On the basis of the knowledge map of historical relics,the design and implementation of the knowledge map search system of historical cultural relics are carried out.The main research contents are as follows:（1）According to the characteristics of rich variety,massive knowledge and heterogeneity in the field of cultural relics,this paper adopts a network model combining convolutional neural network（CNN）and bidirectional long short-term memory-conditional random field（Bi LSTM-CRF）in the process of entity recognition.Extract entity information.And through comparative experiments,it is proved that on the basis of the historical cultural relics data set built by this paper,the improved model Bi LSTM-CNN-CRF entity recognition effect is better than the CRF,Bi LSTM,Bi LSTM-CRF models,and the accuracy rate reaches85.77%,F value reached 84.49%.（2）The construction method of the knowledge map and the main steps of constructing the knowledge map of historical relics are expounded in detail,and finally the knowledge map of historical relics is constructed.It is feasible to provide data support for the final historical cultural relics knowledge map search system.（3）Build a knowledge graph search system for historical cultural relics.This system is developed based on the flask framework,using the front-end and back-end separation mode,the front-end uses the bootstrap framework,and the knowledge graph visualization involved in the search process uses the drawing framework echarts.Finally,through the test of system performance and function,the expected results have been achieved,which proves that the system has achieved the purpose of facilitating users to better understand and query the knowledge of historical cultural relics.