Event Logic Graph Construction for Event Mining

Abstract

Event Logic Graph is a directed and cyclic graph, the nodes in the graph represent events, and the edges represent the logical relationship between events. In essence, Event Logic Graph is a knowledge base of event logic, in order to reveal the evolution law and development mode of the event, we research a method of constructing Event Logic Graph, which describes the logical structural relationship between events by adopting knowledge graph structure. At the same time, in order to describe the event more clearly, we also describe the multi-dimensional attributes of the event. We propose an architecture for constructing Event Logic Graph, which includes text corpus collection, event relationship template construction, event extraction and structured representation, event similarity calculation and fusion, event trigger word extraction and argument extraction model construction, event relationship pair construction, graph database storage, and the architecture is used to construct Serial Event Logic Graph, Causal Event Logic Graph, Conditional Event Logic Graph, Transitional Event Logic Graph, and Concurrent Event Logic Graph.