Emergency Event Web Information Acquisition using Crowd Web Sensors

In the domain of emergent event analysis, it is still a difficult issue to acquire the event information from the Web efficiently. To solve the problem, this paper proposes a crowdsensing-based Web crawler for emergent event analysis. When an emergent event occurs, some web users post event information on the Web with geographical position. These web users can be regarded as crowd sensors. In the proposed method, the crawler takes advantage of the information from these crowd sensors, such as semantic information, geographical information, sentiment information, etc., to get the information of event efficiently. Experimental results show that the proposed method can improve the efficiency of crawlers when compared with universal crawlers both in the period of sparse information and in the period of eruptible information.