Research Of Information Clutter Based On Visual Analysis

Data visualization is the process of presenting data in a graphical form so that it can be easily analyzed and understood.In the information age,data visualization has become an important tool in various fields.However,with the rapid growth of data volume,the limited display space is no longer able to fully present all data,which has an impact on the process of in-depth analysis and understanding of data by users,resulting in the problem of information obfuscation.In addition,different users have different criteria for judging information obfuscation,different data visualization methods produce different forms of information obfuscation,and different problem analysis scenarios define information obfuscation differently,so the current challenge is the lack of accurate information obfuscation quantification tools and unified information obfuscation elimination models.Scatter diagrams and social network diagrams are commonly used in data visualization.Scatter diagrams usually represent the relationship between two data by the position and size of the points,but the problem of data point display occlusion occurs; social network diagrams are used to show the relationship between different entities in the network,but the problem of edge intersection and overlap arises in complex scenarios.The information obfuscation in the 2D view leads to the degradation of the quality of data visualization and can obscure important data,thus affecting the user’s interpretation and analysis of the data.To address the above problems,this paper proposes a visual analysis-based solution that aims to solve the information obfuscation problem in two-dimensional views through multiple visualization techniques.1.For the quantification of information obfuscation,this paper proposes an obfuscation degree model based on the concept of information entropy to evaluate the information obfuscation of the global view in terms of the content,distribution and characteristics of the data,and also designs a visual attention model based on the importance of the individual data in the view to the global.2.For the information obfuscation problem of scatter graph,this paper designs an information obfuscation elimination algorithm and a collision detection-based scatter graph layout algorithm based on the visual attention model,which eliminates the obfuscation of visually obscured data points and allocates the display space to the scatter points in the view according to the size of attention and arranges them in order.3.For the information obfuscation in the network graph,this study establishes a new visual coding method to eliminate the edge-related information obfuscation in the social network graph by using the layout of galaxy diagram,and to show the relationship between the nodes in the network graph more intuitively.Based on the framework of scatter plot and network graph information obfuscation elimination models,this study establishes the Info Clutter Vis visual analysis system to verify the effectiveness of the information obfuscation algorithm.The system is also highly interactive,helping users to analyze in depth.After a comprehensive case study of the system in a realistic data-based scenario,it was confirmed that Info Clutter Vis is effective in solving the information obfuscation problem.