Continuous Similarity Learning with Shared Neural Semantic Representation for Joint Event Detection and Evolution

Table 3

Results of event evolution about different story structure generation algorithms.

Method	Consistent Paths	Best Structures	Repeat Readings
JEDE-Graph	—	6.1784	15
JEDE-Flat	—	6.2996	12
JEDE-Timeline	7.5004	7.1860	12
Story Forest	7.8244	7.2706	11
JEDE-Tree	8.1106	7.8626	8

The bold values highlight the performance of the proposed JEDE-Tree model on several metrics. It can be seen from Table 3 that our model with the tree structure performs better than other baseline models. We have described these results in Section 5.4.