A hybrid deep learning architecture for opinion-oriented multi-document summarization based on multi-feature fusion
作者：
Highlights：
• A novel deep-learning-based method for opinion-oriented multi-document summarization.
• Pre-trained deep-learning-based methods for opinion summary.
• Method comprises word embedding, sentiment, statistical and linguistic knowledge.
• Integrating sentence type, contextual polarity, word sense, sentiment shifter rules.
• Results displayed that the method achieved significant accuracy.
摘要
•A novel deep-learning-based method for opinion-oriented multi-document summarization.•Pre-trained deep-learning-based methods for opinion summary.•Method comprises word embedding, sentiment, statistical and linguistic knowledge.•Integrating sentence type, contextual polarity, word sense, sentiment shifter rules.•Results displayed that the method achieved significant accuracy.
论文关键词：Deep learning,Sentiment analysis,Opinion summarization,Linguistic knowledge,Recurrent neural network
论文评审过程：Received 20 May 2020, Revised 4 December 2020, Accepted 7 December 2020, Available online 10 December 2020, Version of Record 16 December 2020.
论文官网地址：https://doi.org/10.1016/j.knosys.2020.106658
