This article has modeled how to compute veracity of web event via uncertainty based on its features distribution on different kind of confident websites, whose data is big. Some web events have positive effects through facilitating public supervision, while others have negative effects, such as MeiLing Guo event. Our work throws light on timely responding to and understanding popular/emergency web events for individuals and social groups. We propose one method to measure web event veracity through calculating web event uncertainty. In the process of calculating web event veracity, the main contributions of this paper are as follows:
Some event features, which influence on the measurement of veracity in the evolution process of web event, are mined to calculate the veracity of web event.
Relations of features play an important role in computing veracity of web event. The relations include distribution of attributes on webpages and classified webpages on websites. Moreover iterative processes are employed to handle these relations.
Matrix operations are done to confirm that the result of the iteration progress is in coincidence with mathematical model. Finally, experiments are made based on the analysis above, and the results prove that the iterative algorithm is promising.
In all, this paper analyzed web event from its event features to the relation among these features. All the analysis contributes to the computation of web event veracity.