Research On Perception Of Dynamic Visual Coding For Time Series Data

With the rapid development of information society,the explosive growth of data volume and the continuous diversity of data types bring great challenges to human beings,but also create potential opportunities.Around the exploration of data mining,a variety of visualization tools and related technologies have been developed,but with the dynamic changes of time series data in the time dimension,the characteristics of time series data put forward new requirements in the field of visualization.Dynamic visualization can display data information to users in animation,interactive interface,etc.It can improve users' experience and reduce cognitive load.Based on the gestalt psychology and cognitive load theory,this paper studies the basic factors that affect the users' visual perception in dynamic visualization with the goal of designing dynamic visualization of time series data that conforms to the human cognitive characteristics,starting from the user's cognitive mechanism of dynamic visual features.This study combines theoretical analysis and experimental validation,and the results are of great significance to the dynamic visualization design of general time series data.The main contents and innovations of this paper are as follows:(1)Starting from the concept and type of time series data,this paper describes the attribute characteristics of time series data: data scale,event attributes,time attributes.Time series data visualization is divided into static chart visualization and dynamic visualization according to data characteristics.(2)Explain the human visual cognitive model and cognitive load model in visualization,introduce the Gestalt theory related to visual cognition in visualization,and then propose the visual encoding method in visualization.The types of visual channels and dynamic visual variables in visualization are systematically summarized,and their perceptive characteristics in the process of visual coding are analyzed.(3)Focusing on visual coding and the perceptual characteristics of dynamic visual variables,graphic perception grouping experiments are designed based on Gestalt theory to study the separability of different types of dynamic visual coding.Based on the cognitive load theory,the design and experiment explores the cognitive performance differences caused by the duration variable of dynamic visual variables to users.(4)Propose a dynamic visualization design process for time series data.The temporal dataset of cultural development classes from 2011 to 2020 of the National Bureau of Statistics is selected,and the dynamic visualization design is completed by combining the visual coding design method and the dynamic visual variable setting strategy proposed in this paper.