A network-based feature extraction model for imbalanced text data

作者：

Highlights：

• A network-based model is proposed for processing imbalanced text data.

• We design a Polar Layer to combine random walk paths with CNN.

• We introduce an electing strategy to improve the performance of the proposed model.

• The proposed model can well solve the problem of unbalanced text data.

摘要

•A network-based model is proposed for processing imbalanced text data.•We design a Polar Layer to combine random walk paths with CNN.•We introduce an electing strategy to improve the performance of the proposed model.•The proposed model can well solve the problem of unbalanced text data.

论文关键词：Complex Network,CNN,Text Analysis,Imbalanced Data,Random Walk

论文评审过程：Received 5 September 2020, Revised 18 January 2022, Accepted 20 January 2022, Available online 2 February 2022, Version of Record 7 February 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116600