Method and System for information extraction using a self-augmented iterative learning

摘要

An information extraction method and system using autonomous augmented iterative learning is presented. The information extraction method using autonomous augmented iterative learning proposed in the present invention comprises the steps of collecting and storing a knowledge base and a text corpus to generate data for machine learning in which an entity name and a relation name are defined by using a knowledge base and a text corpus. Generating a sentence vector model using a knowledge base and text corpus related to extracting stored information, and generating learning data by linking an entity name and extracting a relationship by using it, performing information extraction using the learning data, and extracting the information Reliability verification step of verifying the similarity and consistency with the existing knowledge base on the result data generated by the method and system, and step of newly reinforcing the knowledge base based on the verified reliability data, iterative and effective learning for information extraction And creating and managing data.