A Method for Automatically Constructing Event Logic Graph from Online News Texts

This paper proposes a framework to address the challenge of extracting critical information from the substantial growth of online news in the Internet age by automatically constructing event logic graph from unstructured news reports. The proposed framework consists of three layers. The data support layer employs crawler technology to gather news from different media platforms and store them in a MySQL database. The data mining layer employs natural language processing technology to analyze the news data and identify events and event relationships, which is utilized to create an event-centered news event logic graph by extracting information. In the visual interaction layer, the Neo4j graph database is utilized to showcase the examples for the news event logic graph. The event logic graph created through online news can enhance users' efficiency in obtaining news event information and aid them in making more comprehensive and scientifically-informed analytical decisions.