Study On Micro-blog Data Visualization Based On Grey Relational Analysis

Association analysis visualization is a nontrivial process of identifying valid,potential,useful,and understandable relational patterns from data.As a communication medium of social networks,micro-blog plays an irreplaceable role in the spread of public opinion.And the emergence of popular micro-blog make more rapid dissemination of information,is the social public opinion guide.Through correlation analysis visualization technology,related links between micro-blog and micro-blog and micro-blog between attributes directly explore,find the relationship of potential,prevention of social public opinion control has an important role,is of great significance therefore the study for correlation analysis of popular micro-blog data visualization.Correlation analysis correlation analysis mining algorithm is very important in the process of visualization,the visualization analysis of associated optimization model in the early stages,the existing relational analysis visualization methods also exist some problems,combining the effective data mining model with good visualization method,in order to obtain valuable results become the focus of research.The main work of this thesis is as follows:1)In order to evaluate the impact of social network influence on the spread of micro-blog micro-blog blogger,combined with the propagation characteristics of micro-blog in social networks in ways and methods of information dissemination,the hybrid algorithm of weighted center linking strength based on the application of the algorithm of weighted center link strength,heuristic algorithm and greedy algorithm combined,is given the user node influence estimation formula to calculate each microblog disseminator influence value,and provide a reliable basis for the evaluation of popular micro-blog attribute weights.The experimental results show that compared with the greedy algorithm and the heuristic algorithm based on node degree,the central weighted link strength algorithm has a significant improvement in accuracy.2)It is difficult to assign the attribute weights objectively after the micro-blog data processing.This thesis presents a method of multi attribute analysis based on gray set.This method combines the influence of bloggers between micro-blog and micro-blog on the property,the property values are defined and classified,the establishment of mining process parameter adjustment strategy,for the selection of key attribute value analysis support.3)In order to solve the problems of the existing association analysis visualization algorithms,a new algorithm based on grey relational analysis is proposed.The algorithm uses the gray correlation analysis theory,the different attributes of the data in the popular micro-blog division,to solve the number of association rules more traditional association analysis method with visual interface disorder and ambiguity problems.The experimental results show that the improved algorithm has good performance,and the proposed visualization has good performance compared with the existing results.