基于可信反馈的微博用户情绪异常预警模型研究

Warning Model of Micro-blog Users' Abnormal Emotion Based on Trusted Feedback

【目的/意义】微博是用户情感发泄的重要渠道,预警模型将有助于发现异常情绪用户,以便及时展开干预。【方法/过程】模型首先利用极性词典和句法序列规则计算微博情感极性程度值,过滤出可疑情绪异常节点;然后利用微博社交网互动关系,计算节点之间的信任值,进一步通过可信反馈对情绪异常节点进行判断。【结果/结论】实验表明,基于序列规则+词典比基于词典的方法对可疑异常情绪用户过滤准确性高,而相比这两种文本挖掘的方法,将可信反馈加入异常情绪判断进一步提高了识别准确度。

[Purpose/significance] Micro-blog is an important way for blogger to vent emotion. The warning model can help find some abnormal emotion users, so that a timely intervening can prevent the occurrence of some extreme behavior. [Method/process]The model calculated micro-blog sentiment degree by polarity lexicon and syntax, and filtered out the suspicious abnormal emotional node. It calculated trust value among micro-blog users through social network interaction, and further to judge the abnormal emotion users through trusted feedback. [ Result/conclusion ] The experiment shows that syntax rule ＆ dictionary-based has a higher accuracy than dictionary-based when filtering the suspicious abnormal emotion users. Compared to the two ways of text mining, trusted feedback can further improve the recognition accuracy.