The design of Public Opinion Analysis System Based on Topic Events

摘要

Public opinion analysis of the current system are generally aimed at one industry or the whole Internet, which led to excessive depth of lack of depth, to the subject of an Internet event information for in-depth analysis of public opinion. Through this research results, can public opinion to a specific event information for in-depth analysis of the multi-dimensional, the use of such public opinion analysis system for Government or large companies found that people focus on specific subject matter, social relations and events that exist within some hidden, Realization of the principle of association rules based on the analysis of social relations, and analyzed by both the support and confidence of the community property of the relevant entities, the module can be more accurately identify potential events related to the page through social entity-relationship of the mining algorithm social relations and can be used as recommendations for Government or corporate decisions, related technical research in the field to have some reference value.