A METHOD AND AN APPARATUS FOR ANALYZING MARKETING INFORMATION BASED ON KNOWLEDGE GRAPHS SUPPORTING EFFICIENT CLASSIFYING DOCUMENTS PROCESSING

摘要

A document classification processing apparatus according to an embodiment of the present invention includes: a classification system definition unit defining classification system information by analyzing a plurality of marketing document data for constructing a marketing specialized knowledge graph; a rule-based filtering unit that performs primary rule-based filtering corresponding to the plurality of marketing document data according to the definition of the classification system information to first remove garbage data; and a machine learning filtering unit configured to perform secondary removal of garbage data by processing machine learning filtering based on a pre-learning model corresponding to the residual marketing document data according to the primary filtering.