Research interest mining method, device and storage medium for scholar clustering

Abstract

The present invention provides a research interest mining method, device and storage medium for scholars clustering. The method includes the following steps: constructing an academic metadata set based on academic data related to scholars from multiple sources; inputting academic metadata as input data into a pre-established In the research interest mining model of , the topic model is sampled to obtain the semantic representation of scholar interest, which includes professional field-topic distribution, topic-English word distribution, topic-Chinese word distribution and topic-scholar distribution; Semantic interest representation performs scholar clustering and obtains scholar clustering results; the research interest mining model shares the same topic distribution for the data of scholars from the same data source and belongs to the same professional field. In the research interest mining model, the professional field-subject distribution is modeled as a Dirichlet distribution, and the subject-English word distribution, subject-Chinese word distribution, and subject-scholar distribution are modeled as multinomial distributions.

Description

technical field [0001] The invention relates to the field of big data technology, in particular to a research interest mining method, device and storage medium for scholars clustering. Background technique [0002] Academic data such as scholars, research projects, and papers all have their own areas of expertise. For example, some data belongs to software engineering and some data belongs to artificial intelligence. The research content of different professional fields is different, and the data of the same professional field often have a common theme distribution. For academic data of scholars, discovering research interests of scholars from academic data and clustering scholars according to their research interests are important for many tasks, such as selecting collaborators for scholars, reviewers for journals, and selection for governments expert. Different from general data, scholarly interest-related academic data has its unique properties. On the one hand, it is...

AI Technical Summary

Problems solved by technology

However, none of these dynamic representation models can track scholars' research interests from academic data

On the one hand, the publication time of papers and project application time are discontinuous, which is not suitable for online processing of academic data

On the other hand, existing batch methods are only suitable for single-source or monolingual data, and thus cannot be directly used to obtain dynamic semantic representations of scholars' research interests.

Images