Design And Implementation Of A Public Opinion Analysis System On The Internet

In recent years, the e-commerce and the social networking services have developedrapidly. The network platform provides convenience for publishing consumers’ productevaluation and the experience of consumption and at the same time, it makes the publicopinions spread explosively. Researches show that consumers are more likely to believethe public opinions rather than marketing advertising, and the public opinions have a hugeimpact on enterprises and the society. In recent years, various types of accidents aboutproduct quality occur in China, and the spread of public opinions has played an importantrole in promoting the spreading of such events. The analysis of public opinions data cannot only help the companies to improve product quality, but also timely detect productsdefects and problems as well as supervise the production enterprises.This thesis, combined with current public opinions data analysis needs, designs andrealizes a public opinions analysis system of good user experiences. In the light of thepublic opinions data characteristics, this system provides users with personalizedmultidimensional analysis services of public opinions data and user-friendly visual userinterface. This thesis consists of three parts of specific work:1) Multidimensional analysisof the public opinions data.2) Free customized analysis of the different types of users.Users can create their own analysis components and information centers and can freelyconfigure the attributes of analysis components, such as the dimensional attributes and theattributes of chart types. And the system analyzes the results according to the attributes ofanalysis components.3) The presentation of visualization results. Because the publicopinions data has the features of diversity and complexity, the results of multidimensionalanalysis will be presented in the form of a visual chart which is very intuitive easy tounderstand.The analysis system of public opinions has achieved the multidimensional analysis ofpublic opinions data and provided a user-friendly visual user interface. This system hasbeen tested and verified and can satisfy the design requirements. The achievements of thisresearch are of great value for public opinions researches.