Document title tree construction method, device, electronic equipment, storage medium, and program

摘要

Problem to be solved: to provide a method for identifying a title of an unstructured document and a method of constructing a document title tree capable of constructing a document title tree, an electronic equipment, a storage medium and a program.How to build a document title treeRule matching the text features of each paragraph in a document to be processed based on predefined rules and paragraph features in a predefined rule, andReturnsDetermining the paragraph level of each paragraph in the document to be processed based on the result of the rule matchingReturnsDetermining the paragraph level of each paragraph in a document to be processed using a machine learning modelBuilding the document title tree for the document to be processed based on the paragraph level of each paragraph.Diagram