中文摘要：

微博的高参性与和广泛用户来源使其成为舆情监测与竞争情报搜集的首要信息来源。首先界定了用户中心化加工的目标，并建立二层复杂网络模型以描述微博系统的结构特征。以中心化为目标，提出了从微博层到用户层的映射机制，输出用户在微博危机事件中交互的社会网络模型。通过对该模型的中心化度测度提出对危机事件中用户的角色划分。最后结合危机案例验证了微博危机事件中心化加工中的网络模型、映射机制和用户划分。

英文摘要：

Micro-Blogging platform becomes the chief information source of public opinion detection and competitive intelligence collec- tion, for the high users participation and wide range of users sources. The article defines the target of centralization of users, also builds the 2 layers complex network model to present the network feature of Micro-blogging system. Aiming at centralization, we propose a mapping method from Micro-Blogging layer to users layer, the output of mapping method is a social network which presents the communication a- mong users in Micro-Blogging crisis event. By the centralization measure of the social-network model, we propose a role partition rule of the users in Micro-Blogging. Finally, we verify the network model, mapping-method and partition of users by a real crisis case.