Information Visualization Technology And Applied Research

Information visualization is the study of visual representations of abstract data to reinforce human cognition. In the field of finance, network communication, biology and business intelligence, information visualization is widely used for assisting the user to analyze the datasets and then to find the laws contained in the data. With the rapid growth of datasets and diversity of data sources, the traditional information visualization technologies are facing with the new challenges. Therefore, research on the technologies of information visualization has great important theoretical significance and practical value.This thesis mainly focuses on information visualization techniques and applications. The methods of visualization and visual analysis are proposed for microblogging data, financial data and statistics data. And, the experiment results have shown that our proposed methods are efficient and accurate.Firstly, an approach of financial data visualization based on the gravitational field clustering is proposed. Self-organizing map is used to classify the raw financial data. Then, the gravitational field theory is used to congregate fold line in each class and meanwhile to set the exclusion between the classes. Moreover, the visualization results are enhanced by setting the opacity and interaction to better analyze the original financial data and give investors some clues. In order to verify the effectiveness of that proposed algorithm, a real world financial data is used. The experimental result shows that the proposed method forms a clear visual clustering result and discovers the variation law of the data. Users can easily select the merit investment value of the company to make investment decisions.Secondly, we propose a novel approach based on ontology for Web information extraction and visual analysis. We sum up four features for information items and induce these features to a group of extraction rules. Then according to a group of mapping rules between elements of ontology and extraction rules, extraction rules are well organized in ontology. According to properties of concept in ontology, the initial result of information extraction is got and then the final result is obtained by simplifying the initial results. Then we combine google map to visual analysis the extracted Web Information, which facilitate the users to understand data mining, find hidden features, relationships, patterns and trends.Thirdly, we proposed a data visualization method based on Particle Swarm Optimization (PSO) for microblogging data, aiming at solving the problem that is hard to find relationships between nodes due to the complex social networking nodes that are either too concentrated or too diversified. First, the microblogging data are divided into subgroup based on relationship between these data. We adopted the PSO algorithm to design objective functions in order to design the layout of the microblogging users’network, distribute the nodes in a more balanced way, and reduce the intersection of the line segment. To verify the effectiveness of that proposed algorithm, we use the data obtained from Sina.com and the Tencent microblogging sites. The experimental results have shown that the proposed method formed a clear visual result and provided a better analysis of relationship among the microblogging users.Finally, we propose a mixed visual analysis method combined with geographic information data to better display temporal and spatial statistical data. The proposed system combines a variety of visualization techniques, such as parallel coordinates, geographic maps, dynamic scatterplots, Treemap and other visualization techniques to enhance the visual results; which ensures that users can observe various indicators data from different angles. At the same time, in order to further analyze the data of interesting, we design and achieve a variety of visualization techniques coordinated with multiple views. The experimental results show that the proposed mixed visualization techniques can help users to discover the variation law of data more easily and offer a fast and convenient visualization tool for statistical data analysis.This thesis explores new ideas and technical methods to solve the current problems faced in the field of information visualization tackling with data mining, visual analysis and graphics hardware technologies. And at the same time, a large number of experimental results verify the effectiveness of the proposed methods.