Does deep learning help topic extraction? A kernel k-means clustering method with word embedding

作者：

Highlights：

• A novel topic extraction method incorporated with a kernel k-means model and a word embedding model.

• The incorporation of word embedding techniques in data pre-processing for topic extraction.

• A polynomial kernel function-based k-means model for effectively conducting bibliometric data-oriented topic extraction.

• Empirical insights into both overlapping and diverse research interests among three top-tier bibliometric journals.

摘要

•A novel topic extraction method incorporated with a kernel k-means model and a word embedding model.•The incorporation of word embedding techniques in data pre-processing for topic extraction.•A polynomial kernel function-based k-means model for effectively conducting bibliometric data-oriented topic extraction.•Empirical insights into both overlapping and diverse research interests among three top-tier bibliometric journals.

论文关键词：Bibliometrics,Topic analysis,Cluster analysis,Text mining

论文评审过程：Received 23 January 2018, Revised 7 September 2018, Accepted 7 September 2018, Available online 24 September 2018, Version of Record 24 September 2018.

论文官网地址：https://doi.org/10.1016/j.joi.2018.09.004