Emotion Information Visualization And Richness Analysis Based On Enriching Mental Engineering

Information visualization (InfoVis) was an important subject for text data mining, also as the important method to get useful knowledge from single information data for people. And the way which is called data mining was exactly the approach to extract potential and valuable knowledge for user from massive data. Visualization means a process of transferring data or knowledge to certain visual form via computer processing, then implement the human's visual sense to use its natural ability to recognize these information faster, it also means it will achieve the high efficient transformation through improving people's understanding of knowledge.In this paper, firstly, by establishing the structure for the implementation of visual model to realize a method of structural organization of thinking, and then the Chinese emotion as the original data of the XML corpus Ren_CECps 1.0 documentation set for further manual processing and filtering the data available source, the last the open source, scalable visualization software development kit prefuse will be processed the text with emotional information, and aiming at different emotional information function, we achieved visualization interface based on emotion information marking structure, time sequence, force-oriented, scattered distribution and visualization Elements on 3D visual interface.At emotion data analysis aspect, we mainly use the criteria fraction in psychological statistics to measure the differences between the average concentration and tests in a test set which including emotion information(non-zero value), then we will have a result on comparison way of single text and the overall average value. At the same time we implement the quality identification method based on chi-square distribution to judge the significant diversity between target text set in a certain period of time and the standard set, in order to understand whether the author's motional change belong to the custom "Normal" scope. It also provided a feasible approach for processing similar text samples in our future research.