Method and system for extracting Chinese key phrases in scientific and technological innovation field by utilizing semantic features

Abstract

The invention discloses a method and a system for extracting Chinese key phrases in the scientific and technological innovation field by utilizing semantic features. According to the method, Chinese stop words and a stop mode library are constructed by mining corpus features of Chinese scientific and technological innovation documents, so that high-performance filtering of invalid information is realized; in addition, various key phrase extraction algorithms are quantitatively evaluated and analyzed by means of domain expert labeling, so that an algorithm model more suitable for domain cognition is selected, and multiple statistical rules are used for filtering to improve phrase extraction performance; and the structural characteristics of the document are further utilized to carry out vector space embedding representation on the topic semantics of the document, and the semantic similarity between the extracted phrases and the topic of the document and the semantic importance degree of the phrases are comprehensively utilized to carry out calculation and ranking so as to finish further screening of the key phrases. The method can support various downstream tasks and applications, including scenes of scientific and technological innovation field knowledge graph construction, scientific and technological innovation document semantic retrieval, scientific and technological innovation entity accurate search and the like.

Description

technical field

[0001] The invention relates to the fields of computer systems, big data, artificial intelligence, knowledge map construction, natural language processing, etc., and specifically relates to a method for extracting key phrases in the field of scientific and technological innovation using semantic features. Background technique

[0002] Traditional key phrase extraction in the field of scientific and technological innovation relies on manual operations and requires relevant staff to have rich relevant professional knowledge. If the extracted key phrase field does not match the personnel knowledge field, it will often lead to errors in judging and extracting phrases. The key phrases extracted manually are prone to problems such as incompleteness, lack of detail, untimelyness, and inconsistency with the direction of objective needs. Therefore, the traditional artificial key phrase extraction method has defects such as heavy workload, low efficiency, high error ra...

Images