Research And Implementation Of Urban Catering Consumption Data Visualization

With the rapid development of the era of big data,the penetration of the catering industry and Internet technology has gradually increased,and the online life service platform has become the main channel for consumers to obtain catering information.However,most of the online life service platforms are mainly oriented to operators and supplemented by the needs of consumers.At the same time,the massive data and information are complicated and there are untrue information such as advertisements,which makes it very difficult for users to choose a dining place.Visual analysis technology aims to mine effective information from a large amount of complex and incomplete data.Therefore,this dissertation combines catering data and visualization technology to design and implement a consumer-oriented urban catering consumption data visualization system to help users understand the real situation of restaurants better and assist decision-making.The main work is as follows:1.A data collection scheme designed based on the Scrapy crawler framework.By analyzing the web page structure and data characteristics of the online life service platform,a collection scheme for dynamically generating a mapping dictionary is designed to effectively capture the data.2.A multi-dimensional analysis hierarchical ring graph visualization method for comment text is proposed.It mines the key words of the text through topic extraction technology,and uses machine learning algorithms to classify the sentiment of the review text.While retaining the time series characteristics of the data,it can simultaneously analyze the emotional dimension,vocabulary dimension,and time series dimension data information.Verified by experiments on real data,this visualization method can highlight the key information of the comment text,and have a good visualization effect.3.A visualization method of parallel coordinate dimension scaling based on binding effect is proposed.The data is dynamically obtained through visual interaction technology to achieve dynamic refresh of coordinate value fields and dynamic generation of coordinate dimensions,and then the combination of parallel control point groups and virtual binding axes are used to match new data points and coordinate value fields,and data curves are drawn to complete the coordinates dimension scaling.Verified by experimental on public datasets,parallel coordinates that support dimension scaling can help users observe and analyze data attributes more intuitively.4.A urban catering consumption data visualization system is designed and implemented.According to the data characteristics,six functional modules,information input,map visualization,comment text visualization,catering consumption data visualization,store details visualization and catering data overview visualization are designed and implemented,and the interaction between each functional module is developed,which may help user make dining decisions that meet the requirements.