中文摘要：

为研究社会网络用户情感行为，以新浪微博数据为基础，对用户微博内容情感进行分级，进而对微博用户群体情感行为进行分析，建立了分级别情感发帖模型，并进行了仿真实验．通过分析发现用户在表达某一级别情感的发帖量均服从幂律分布，且幂指数随着情感级别趋向平和而增加．大部分用户通过微博表达情感时较为平和，需要表达激烈的情感时，用户参与比例会减小．在级别情感发帖模型的建立过程中，考虑到了发帖人受到周围情感环境因素的影响以及自身情感的随机性变化，并通过模型仿真的结果验证了分级别情感发帖量服从幂律分布以及幂指数的变化趋势．

英文摘要：

In order to study the social network users＇ emotional behavior, the massive sina micro-blog- ging data is graded by emotional stage, and the group emotional behavior is analyzed as well. An emo- tional stage posting model is established and model simulation is done. Analysis shows that the posting a- mount of users obeys power-law distribution at a certain emotional stage, and the exponent increases as the emotional stage tends to calm. Most of posts are peaceful when users express emotion through the mi- cro-blogging, and the user participation rates will also deerease when the intense emotions need to be ex- pressed. In the emotional stage posting model, the impact of the around emotional environmental factors of the poster is considered, and also the randomly changing of poster＇ s own emotion. Finally the previous experiments are verified by model simulation.