计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (7): 233-240.doi: 10.11896/jsjkx.240600144

• 人工智能 • 上一篇    下一篇

结合评价对象信息的评论摘要研究

孔银玲, 王中卿, 王红玲   

  1. 苏州大学计算机科学与技术学院 江苏 苏州 215006
  • 收稿日期:2024-06-24 修回日期:2024-09-20 发布日期:2025-07-17
  • 通讯作者: 王红玲(hlwang@suda.edu.cn)
  • 作者简介:(20225227032@stu.suda.edu.cn)
  • 基金资助:
    国家自然科学基金(61976146);国家自然科学基金青年科学基金(61806137)

Study on Opinion Summarization Incorporating Evaluation Object Information

KONG Yinling, WANG Zhongqing, WANG Hongling   

  1. School of Computer Science and Technology, Soochow University, Suzhou, Jiangsu 215006, China
  • Received:2024-06-24 Revised:2024-09-20 Published:2025-07-17
  • About author:KONG Yinling,born in 2000,postgra-duate,is a member of CCF(No.L5741G).Her main research interests include text summarization and opinion summarization.
    WANG Hongling,born in 1975,assistant professor,is a member of CCF(No.14272M).Her main research interests include natural language processing and information retrieval.
  • Supported by:
    National Natural Science Foundation of China(61976146) and Young Scientists Fund of the National Natural Science Foundation of China(61806137).

PDF (PC)

19

摘要： 评论是消费者对商品评价和反馈的一种文本形式。评论摘要是指对评论进行提取和压缩,形成能够概括评论信息的短文本。目前,评论摘要任务大多只关注评论的文本序列,忽略了评论中的方面、意见短语和情感极性等相关评价对象信息。因此,提出了一种基于T5模型(Text-to-Text Transfer Transformer),结合评价对象信息的评论摘要方法。该方法首先利用T5模型对评论摘要任务进行建模,通过注意力机制学习评论文本中的上下文信息,生成包含核心语义的摘要文本;然后提取摘要文本中的评价对象信息,并将其作为评论摘要任务的辅助信息;最后利用少样本数据对模型参数进行特异性调整,进一步改善摘要的效果,从而生成高质量的评论摘要。实验结果表明,在酒店评论数据集SPACE和产品评论数据集OPOSUM+上,该方法相较于基准模型在ROUGE评价指标上均有显著提升。

关键词: 评论摘要, T5模型, 评价对象信息, 少样本数据, 注意力机制

Abstract: The opinion is a written form of consumer evaluation and feedback on a product.Opinion summarization refers to extracting and compressing a review to create a concise text that summarizes the information contained in the review.Currently,most opinion summarization tasks focus solely on the review text,without considering the evaluation object information within the review,such as aspects,opinion phrases,and sentiment polarities.Therefore,this paper proposes an opinion summarization me-thod based on the T5 model(Text-to-Text Transfer Transformer) incorporating the evaluation object information.The method first utilizes the T5 model to represent the task of opinion summarization.It learns contextual information from the review using the attention mechanism and generates a summary that encapsulates the core semantics.Then,it extracts the evaluation object information from the summary as an auxiliary task of opinion summarization.Finally,the model parameters are fine-tuned using a limited sample of data,which further enhances the summary generation process,resulting in a high-quality summary.Experimental results on both the hotel review dataset SPACE and the product review dataset OPOSUM+,show that the proposed method has a significant improvement in the ROUGE evaluation metrics compared to the baseline models.

Key words: Opinion summarization, T5 model, Evaluation object information, Limited sample data, Attention mechanism

中图分类号: 

  • TP391
[1]RAFFEL C,SHAZEER N,ROBERTS A,et al.Exploring the limits of transfer learning with a unified text-to-text transformer[J].Journal of Machine Learning Research,2020,21:5485-5551.
[2]CHU E,LIU P J.Meansum:A neural model for unsupervisedmulti-document abstractive summarization[C]//Proceedings of the 36th International Conference on Machine Learning.2019:1223-1232.
[3]BRAZINSKAS A,LAPATA M,TITOV I.Unsupervised opinion summarization as copycat-review generation[C]//Procee-dings of the 58th Annual Meeting of the Association for Computational Linguistics.2020:5151-5169.
[4]ISO H,WANG X,SUHARA Y,et al.Convex aggregation foropinion summarization[C]//Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing.2021:3885-3903.
[5]ISONUMA M,MORI J,BOLLEGALA D,et al.Unsupervised abstractive opinion summarization by generating sentences with tree-structured topic guidance[J].Transactions of the Association for Computational Linguistics,2021,9:945-961.
[6]AMPLAYO R K,LAPATA M.Unsupervised opinion summarization with noising and denoising[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics.2020:1934-1945.
[7]MENG X,WEI F,LIU X,et al.Entity-centric topic-oriented opinion summarization in twitter[C]//Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Disco-very and Data Mining.2012:379-387.
[8]SHEN M,MA J,WANG S,et al.Simple yet effective synthetic dataset construction for unsupervised opinion summarization[C]//Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics.2023:1853-1866.
[9]SUHARA Y,WANG X,ANGELIDIS S,et al.Opiniondigest:A simple framework for opinion summarization[C]//Proceedings of the 58th Annual Meeting of the Association for Computa-tional Linguistics.2020: 5789-5798.
[10]VASWANI A,SHAZEER N,PARMAR N,et al.Attention is all you need[C]//Proceedings of the 31st International Confe-rence on Neural Information Processing Systems.2017:6000-6010.
[11]ZHANG M,ZHOU G,HUANG N,et al.Asu-osum:Aspect-augmented unsupervised opinion summarization[J].Information Processing and Management,2023,60(1):103138.
[12]SILEDAR T,MAKWANA J,BHATTACHARYYA P.Aspect-sentiment-based opinion summarization using multiple information sources[C]//Proceedings of the 6th Joint International Conference on Data Science & Management of Data(10th ACM IKDD CODS and 28th COMAD).2023:55-61.
[13]LERMAN K,BLAIR-GOLDENSOHN S,MCDONALD R T.Sentiment summarization:Evaluating and learning user preferences[C]//Proceedings of the 12th Conference of the European Chapter of the Association for Computational Linguistics.2009:514-522.
[14]WANG K,WAN X.Transsum:Translating aspect and senti-ment embeddings for self-supervised opinion summarization[C]//Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing.2021:729-742.
[15]BHASKAR A,FABBRI A R,DURRETT G.Prompted opinion summarization with GPT-3.5[C]//Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics.2023:9282-9300.
[16]BROWN T B,MANN B,RYDER N,et al.Language Models are Few-Shot Learners[J].arXiv:2005.14165,2020.
[17]AMPLAYO R K,ANGELIDIS S,LAPATA M.Aspect-controllable opinion summarization[C]//Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing.2021:6578-6593.
[18]GLIWA B,MOCHOL I,BIESEK M,et al.SAMSum Corpus:A Human-annotated Dialogue Dataset for Abstractive Summarization[J].arXiv:1911.12237,2019.
[19]LEWIS M,LIU Y,GOYAL N,et al.BART:denoising sequence-to-sequence pre-training for natural language generation,translation,and comprehension[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics.2020:7871-7880.
[20]LIN C Y.Rouge:A package for automatic evaluation of summaries[C]//Text Summarization Branches Out.2004:74-81.
[21]ERKAN G,RADEVD R.Lexrank:Graph-based lexical centrality as salience in text summarization[J].Journal of Artificial Intelligence Research,2004,22:457-479.
[22]ANGELIDIS S,AMPLAYO R K,SUHARA Y,et al.Extractive opinion summarization in quantized transformer spaces[J].Transactions of the Association for Computational Linguistics,2021,9:277-293.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发