Quantitative Rearch On The Perception Difference Of Ensemble Coding On The Information Visualization Interface

To extract the hierarchical structure and associated attributes of complex information on the information visualization interface,the visual system needs to be able to integrate multiple meta-information simultaneously.As an effective visual statistical method to extract information aggregation attributes,ensemble coding is a feasible algorithm to provide overall perception.And under the background of the increasingly complex information display dimension and data dimension,the ensemble coding perception result of of overlapping visual features is the premise of user decision accuracy.Therefore,perceptual quantization mechanism of ensemble coding provides a more systematic and comprehensive scientific basis for the design and evaluation of complex information system interface.This research is oriented by the needs of information visualization interface,with the purpose of building a task-driven ensemble coding perceptual difference mechanism,and takes the relevant research methods of cognitive psychology and design science as the technical means.This research could finally find out the information entropy mode of human visual perception,and forms the design evaluation methodology of information visualization based on different ensemble tasks.The main contents and innovation points of the paper are as follows:(1)Researched on the perception basis and the forming factors of perception difference.And the summary and analysis of the information coding design end and the information display end of the visual interface were discussed;and the generation mechanism of the ensembe coding,the hierarchical classification of the ensemble perception,the perception strategy of the ensemble coding and the quantitative perception basis of all the coding were analyzed.It layed a theoretical foundation for the research on the quantization perception difference mechanism of ensemble coding on the information visualization interface and provided a basis for the experimental design methods involved in quantization perception experiment.(2)Information visualization tasks,interaction tasks and analysis tasks were classified.Based on the classification method of operational objects,the information dimension was integrated.The mapping relationship between different information dimensions and ensemble tasks was built,and the ensemble task domain for information visualization interface was finally created.It layed a foundation for establishing the representational relations between different ensemble tasks and ensemble coding.(3)In order to find out the information entropy pattern of human visual perception and built the fitting relationship between physical changable and perception changable of ensemble coding for different task-driven information interfaces,the perception difference mechanism of ensemble coding driven by recognition task,pattern judgment task and magnitude summary task were studied.(4)According to the coding principles and interface design rules obtained from different information visualization tasks,an ensemble coding decision set for visual design was formed.Based on this set,a relevant interface example was designed to verify the effectiveness and rationality of the decision set.And the method of comprehensive evaluation,which combines weighted average operator and VIKOR,also was embedded in the database of interface comprehensive evaluation.Finally,the methodology of information visual design evaluation of relative system was formed.