Method for analyzing and sequencing academic influence of theme literature in citation database

Abstract

The invention discloses a method for analyzing and sequencing academic influence of theme literature in a citation database. The method includes performing theme retrieval in the citation database to acquire metadata information of the theme literature and citation relation among the theme literature; according to the acquired metadata information of the theme literature and the acquired citation relation among the theme literature, creating a theme citation network; on the basis of the citation network, building a factor graph model containing various literature academic influence factors; using a circulating maximum sum algorithm for approximate reasoning on a factor graph to determine marginal probability values of all variables in the factor graph; according to a descending order result of the marginal probability values, acquiring and outputting an academic influence sequence of all theme literature. By the method, sequencing of the academic influence of literature with a theme given by a user can be realized.

Description

technical field [0001] The invention relates to the technical field of citation analysis, in particular to a method for analyzing and sorting the academic influence of subject documents in a citation database. Background technique [0002] Sorting or ranking the academic influence of scientific and technological literature is beneficial for scientific and technological workers to select high-quality literature, obtain authoritative knowledge, and improve the efficiency of scientific and technological work and the quality of scientific research results. The analysis and evaluation of academic influence of literature belongs to the technical field of citation analysis in Bibliometrics, and has a wide range of applications in the fields of library and information science. The needs of scientific and technological workers (or "users") to search literature and expect to obtain high academic impact literature often start from the user's interested topic. Therefore, the academic i...

AI Technical Summary

Problems solved by technology

However, there are problems: 1. The correlation between the literature and the retrieval topic cannot fully reflect the academic influence of the literature; 2. The number of citations cannot fully reflect the academic influence of the literature. Citing documents cannot be considered to have the same quality or influence, and documents that have been published for a long time are more likely to obtain high citation counts, and citation counts in particular cannot measure the academic influence of a document on a specific topic, because when counting citation counts Literature citations do not distinguish between topics; 3. The influence of literature source publications (for example, journal impact factors, etc.) cannot simply be used to evaluate the influence of a single literature

However, its main shortcomings are: 1. The ranking of academic influence of literature has nothing to do with the topics of interest of users; 2. Pure link analysis that only considers the citation relationship of literature cannot reasonably evaluate the academic influence of literature

However, its main shortcomings are: it does not support the importance ranking of given user query topics, and the factors of academic influence of literature considered are relatively simple

However, its main shortcomings are: it does not support the importance ranking of given user query topics, and the factors of academic influence of literature considered are relatively simple

However, its main deficiencies are: 1. It does not provide a method for sorting topic literature on topics of interest to users; 2. The ranking of papers is based on the so-called "popularity" ranking based on link analysis, rather than topic literature that comprehensively considers multiple influential factors Ranked by academic influence

But its main disadvantages are: 1. Document ranking has nothing to do with user interest topics; 2. Only considering the number of citations can not produce a reasonable academic influence ranking of documents

However, the common defects of these two sorting methods are: 1. Document sorting is irrelevant to user interest topics; 2. On the basis of document citation relationship, only the factor of document publication time is considered, which is an idealized random walk model, but in practice, scientific and technical workers will also consider other factors when selecting documents, such as: the influence of the source publications of the documents, the number of citations of the documents, etc.

However, the main disadvantages of this improved method are that it does not consider the topic relevance of literature citations, and does not support the ranking of literature academic influence on user-specified topics.

However, its shortcomings are: it does not consider the topic relevance of literature citations, and does not support the ranking of literature academic influence on user-specified topics

However, its disadvantages are: 1. The relative importance of literature can only be sorted on the topics obtained by probabilistic topic modeling analysis, and the academic influence ranking of literature on topics of interest given by users cannot be realized; 2. Probabilistic topic modeling Using text analysis technology, there are too many topics generated (hundreds or even thousands) and the topics are often unreasonable or unrealistic, which is difficult to be practical; 3. The importance ranking of documents only considers the factor of topic relevance, so the ranking results are not enough Reasonable

However, the main drawbacks of this method are: 1. The literature influence ranking can only be performed on the topics obtained by the probabilistic topic modeling analysis, and the academic influence ranking of the literature on the topic of interest given by the user cannot be realized; 2. The considered influence The factor is only the citation relationship between documents

[0019] In addition to their own shortcomings, all the methods listed above have a common defect: they fail to make full use of the subject search function of existing citation databases to collect relevant literature and various influential factor data on topics of interest to users, and Systematic modeling and scientific analysis of many factors

[0020] Therefore, it is necessary to propose a more effective method for analyzing and sorting the academic influence of subject literature, so as to make full use of the subject retrieval function of existing citation databases to collect relevant documents and various influential factor data on topics of interest to users. And use the probabilistic modeling and inferential calculation functions of the factor graph to systematically model and scientifically analyze many academic influence factors, so as to overcome the existing literature academic influence ranking methods that cannot produce ranking results for the topics of user interest or have not comprehensively considered multiple factors. The defects of unreasonable and inaccurate sorting results due to various academic influence factors, in order to improve the user's scientific and technological work efficiency and the quality of scientific research results

Images