Research On Method Of Spatio-Temporal Multivariate Symbolization Based On Metaphor Theory

In recent years,with the rapid development of information technology and network communication technology,as well as the improvement of information infrastructure,global data has shown an explosive growth,and people have entered the era of Big data.Our life,work and thinking have changed in the information storm brought by big data.Big data thinking focuses more on full sample data rather than local data,and more on correlation rather than causation.Among all kinds of big data,spatio-temporal data is a typical type of big data,which has the characteristics of spatial dimension,time dimension and attribute dimension.Visualization of spatio-temporal data is an effective method to reveal the internal characteristics of data and assist researchers in related fields to conduct research and analysis.However,spatio-temporal big data has a large amount of data and a large number of variables,and do not have a unified dimension characteristics of spatio-temporal data analysis has brought a lot of difficulty,also poses challenges for existing visualization methods: both need to be able to meet the needs of the whole sample correlation analysis,and large amount of data need to be able to avoid or mitigate its visual clutter and overplotting,to support the exploratory analysis of large data.Current methods of data visualization include direct visualization,summaries visualization and patterns visualization.Among them,summaries visualization and patterns visualization are both used to process the data and only show some part of the data,which cannot meet the needs of the correlation analysis of the whole sample.Only direct visualization is to draw the data directly,which can make the data speak and meet the needs of full-sample correlation analysis.However,when directly characterizing large-scale data,it is difficult to recognize the graphics caused by visual clutter and overplotting.In view of the above problems,this paper proposes a direct visualization method by means of metaphor,which is specifically manifested in the design of metaphor symbols,to meet the requirements of whole-sample correlation analysis,avoid or alleviate visual clutter and overplotting,and support the exploratory analysis of big data.Therefore,this paper focuses on the symbolization method of spatio-temporal multivariable based on metaphor theory.The main research contents and achievements are as follows:(1)Analysis of spatio-temporal multivariable symbolic cognitive process based on metaphor theory.Due to the cognitive differences between map designers and map readers,this paper analyzes the cognitive process of spatio-temporal multivariable symbolization based on metaphor theory from the perspectives of designers(construction of metaphor)and map readers(interpretation of metaphor).(2)Symbolized design of spatio-temporal variables based on metaphor theory.According to the cognitive process analysis,design principle summary,a metaphor-based spatio-temporal multivariable symbolization method is designed.According to two specific spatio-temporal multivariable cases,according to the different data types,the appropriate mapping method of metaphor body and variable is selected for symbolization.(3)Usability analysis was carried out to evaluate the designed metaphor-based spatio-temporal multivariable symbolization method by combining eye movement experiment and questionnaire.Efficiency,effectiveness and cognitive burden were used as evaluation indexes and compared with the classical or conventional symbolization method.Based on the same reading task,each group invited 25 valid subjects to perform the reading experiment.