Apriori Algorithm-Based Association Rules of Event Language Expression Discovery

Event is the basic unit of knowledge representation, when human beings communicate with each other. Although event is objective in the real world, when it is described by voice, characters and other means, event has some language features. We treated event as the basic unit of knowledge representation for Chinese news texts. We pre-processed event data in the CEC2.0 corpus from positional feature, Chinese POS tagging feature and other aspects. And then, we proposed minding core word association rules for event language expression from CEC2.0 corpus with Apriori algorithm and optimized them. we detected event elements from Chinese news texts with the mined association rules, including action, object and time. The value of F1 is 60.7%.