Information Visualization Design And Algorithm With Style Features

Information visualization is a rapidly growing field of research,which is originated from the mutual promotion and development of computer graphics,artificial intelligence,scientific visualization,and user interface design.Information visualization,as an interdisciplinary research field,combines the theories and methods of computer graphics,human-computer interaction,scientific visualization,data mining,image technology,cognitive psychology,and so on.With the rapid development in recent years,information visualization has become an important research direction in computer science.It aims to study the presentation and analysis of large scale of non-numerical information resources.Information visualization is dedicated to the creation of the means and methods of conveying abstract information visually,so that users can witness,explore,and immediately understand a large amount of information.One of the major challenges in the research of information visualization is the explosion of information produced with the arrival of the "big data" era.Because the data is large,multidimensional,multi-source,heterogeneous,incomplete,inconsistent,inaccurate,if we want to obtain knowledge from mass data by using information visualization,it needs the aids of knowledge related to machine learning,data mining and natural language processing techniques,and so no and,more importantly,we have to make more effect in information visualization and design style of algorithm.This paper first analyzes the current development of information visualization and describe an application exploration of visual analysis.Then we comprehensive research the edge bundling technology.Finally we put forward an approach to bundling edges of big graphs with labels.Graphs depicted as node-link diagrams are widely used in visualizing relational data sets.Visual clutter becomes problematic with big node-link diagrams.Researchers have recently proposed various approaches to bundling edges to reduce visual clutter.None of the edge bundling approaches,however,addresses the issue of labeling in the visualized graph.This paper presents an approach to bundling edges of big graphs,possibly over geographical maps with location labels such as cities and states.Our approach routes edges via flexible control points to leave space for labels to be clearly visible.We demonstrate the effectiveness of our approach on two commonly used datasets.