Research hotspot analysis method based on expert paper big data

Abstract

The invention discloses a research hotspot analysis method based on expert paper big data, which comprises the following steps: S1, carrying out paper search by taking a knowledge database as a data source according to keywords, and capturing open data such as a paper title, publication time, an author and a data source; S2, according to the downloaded thesis title, performing word segmentation processing, deleting structural vocabularies such as continuous words, mesons and pronouns, removing vocabularies of verbs and word properties of adjectives, only retaining nouns, and obtaining a hot word list; S3, deleting daily frequently-used words in the hot word list through a professional paper daily frequently-used word corpus matching and removing method to obtain a professional word list; S4, performing word frequency analysis on the professional vocabulary list, arranging the professional vocabulary list in a word frequency descending order, and selecting first hundreds of bits of datato obtain a professional vocabulary word frequency general table; and S5, adding year data, listing a word frequency score table according to different years on the basis of the professional vocabulary word frequency general table, and obtaining a change trend of a focus point with time as an axis.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a research hotspot analysis method based on big data of expert papers. Background technique [0002] HowNet has its own search engine, which can be searched according to title, author, keywords, etc., and the search results can also be exported and shared with software such as Note express. CiteSpace, a Java application program used to analyze and visualize co-citation networks, can analyze the development process and structural relationship of scientific knowledge. After exporting data such as titles from HowNet, it can complete scientific knowledge genealogy analysis such as keyword analysis and author relationship analysis. . [0003] However, the defect of HowNet’s existing search engine is that it needs to be manually clicked every time. For data analysis, this working method is undoubtedly very laborious. The amount of manually selected and downloaded data is limi...

AI Technical Summary

Problems solved by technology

[0003] However, the defect of HowNet’s own search engine is that it needs to be manually clicked every time. For data analysis, this working method is undoubtedly very laborious.

The amount of manually selected and downloaded data is limited, and usually the labor cost does not allow to download all the data, making it difficult to have an overall and comprehensive analysis of the data.

Moreover, its analysis function is limited to the memory and sorting of keywords entered by users, and there is no deeper analysis function

CiteSpace is still based on manual operation, the efficiency is low, and incomplete data will cause analysis errors, and it cannot solve professional analysis problems

Images