基于网络搜索的事件分析研究——以2008年金融危机为例

Event Analysis Research Based on Internet Search——Take Financial Crisis in 2008 as an Example

网络搜索是一类常见的用户信息行为,通过其可以体现大众的关注点,这为事件分析提供了新的视角。论文构建了一个基于网络搜索的事件分析框架,涵盖了事件相关网络资讯的采集、事件静态词表构建、事件动态演化分析、搜索数据的统计挖掘等主要环节,以2008年美国金融危机事件为例,探讨了网络搜索与金融危机传导的关系,基于统计学方法得出了中美金融危机传导的时延,提出了相应的建议策略。并结合实例验证了该事件分析框架的合理性、有效性和扩展性。

As a common type of user information behavior, Internet search provides a brand new perspective for event analysis by reflec-ting public concerns. This paper builds a framework of event analysis upon internet search, covering the main parts such as the event-rela-ting website information collection, event static vocabulary building, event analysis of dynamic evolution, and data mining of search statis-tics. Using the case of U. S. financial crisis in 2008, the paper explores a correlation relationship between internet searching and financial crisis contagion, draws a conclusion of time lag of the financial crisis transmission mechanism from U. S. to China based on statistical methodology, and proposes the corresponding strategies. Rationality, effectiveness and expandability of the proposed analysis framework are verified with examples and evidences at last.