A new topic modeling based approach for aspect extraction in aspect based sentiment analysis: SS-LDA

作者：

Highlights：

• Aspect based sentiment analysis is made on Turkish smartphone reviews.

• A new aspect extraction method is proposed.

• An adaptation of Latent Dirichlet Allocation for short text is proposed.

• Proposed method can be used on also other languages with some minor adaptations.

• Competitive results are obtained on SemEval-2016 Turkish data set.

摘要

•Aspect based sentiment analysis is made on Turkish smartphone reviews.•A new aspect extraction method is proposed.•An adaptation of Latent Dirichlet Allocation for short text is proposed.•Proposed method can be used on also other languages with some minor adaptations.•Competitive results are obtained on SemEval-2016 Turkish data set.

论文关键词：Sentiment analysis,Opinion mining,Aspect based sentiment analysis,Aspect extraction

论文评审过程：Received 12 March 2020, Revised 2 November 2020, Accepted 2 November 2020, Available online 16 November 2020, Version of Record 24 January 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114231