Theoretical Research And Technical Implementation Of Semantic Map

Semantic map refers to an electronic map based on traditional geographic information and adding semantic information to geographic objects.Compared with traditional electronic maps,semantic maps can provide richer information,and can provide more valuable search results for users' searches,while also helping to reduce search complexity.This article focuses on providing useful semantic information for traditional GIS objects.The paper chooses POI(Points of Interest)in the open source OSM(Open Street Map)as the geographic objects to be studied,such as museums,commercial centers,and scenic spots.,Research using NLP(Natural Language Processing,natural language processing)tools to extract semantic information from POI-related social data(user comments,reviews,etc.),and enrich the corresponding geographic objects with the resulting semantic information,such as defining new semantic tags(Key-value pairs).The paper first selects the microblog data related to the POI as the data set,and uses the jieba Chinese word segmentation tool to extract the keywords of the data set to provide candidate keyword semantic tags for the POI;then select the data set suitable for the research of this paper to train bert Chinese sentiment classification Model,and use the deep learning model obtained by training to classify the bipolar emotion of the POI data set,and provide satisfaction labels for the POI;finally,add new keywords and satisfaction labels to the specified POI according to the OSM label specification.In the paper,Sanya is selected as the POI,which is used as an example of the above research and technical realization.The paper selects a small sample of Weibo data to complete the keyword extraction and sentiment classification tasks,and obtain satisfactory results.Keywords provide users with information related to POI,intuitively express the consensus impression of most users on the location,and can extract the location and name most relevant to POI;satisfaction shows the acceptance of the location from the data The degree of welcome provides data support for evaluating the most popular restaurants,hotels,and attractions.In addition,the paper also studied the method of extracting specific Weibo data by relationship.Based on the method of using Weibo user credit score to evaluate the user's contribution,the idea of cloud relationship extraction data set was proposed,and the subsequent related event extraction and semantic map RDF(Resource Description Framework)works.Finally,the paper introduces semantic description into OSM,and provides semantic support for OSM's semantic query and semantic reasoning;extracting Weibo data through natural language processing to enrich semantic maps can not only make POI geographic information more complete,rich and complete Objective information;and cut into emotional evaluation,adding a subjective impression to the POI.