基于pu-learning的同行评议文本情感分析

doi:10.3778/j.issn.1002-8331.2108-0341

摘要/Abstract

摘要： 最近几年逐渐出现了对同行评议文本情感分析的研究，包括通过同行评议文本预测审稿人的推荐状态的任务。现有模型融入了论文本身或摘要信息，采用神经网络学习论文或摘要的高层表示，结合同行评议文本预测审稿人的推荐状态，这使得模型变得非常复杂的同时结果却没有实质性的提高。为此，提出了OSA机制来提高情感分析模型中对观点句的关注度。具体来说，采用pu-learning从同行评议文本的前[N]个句子中学习观点句的特征，使每一个句子都得到一个观点句权重，将其应用于情感分析模型的倒数第二层，由此得到最终的预测结果。在ICLR 2017—2018数据集上使用不同的情感分析模型对OSA进行了评估，实验结果验证了OSA的高效性，并在两个数据集上取得了优异的性能。

关键词: 同行评议, 情感分析, pu-learning, 数据挖掘

Abstract: There have been some researches on the sentiment analysis of peer review text, including the task of predicting the overall recommendation through a peer review text written by reviewer for a submission. Existing works integrate the embedding of the paper or abstract, utilizing neural network to learn the high-level representation of paper or abstract and review text to predict reviewer’s overall recommendation, which make the algorithm very complicated but the effect is not substantially improved. To solve this issue, a mechanism called OSA（opinionated sentence attention） is proposed to make opinionated sentences get more attention in sentiment analysis model. Specifically, this paper employs a positive-unlabeled learning method to learn opinionated sentence features form Top-N sentences of peer review texts so that every sentence of all review texts gets a opinionated weight, then these weights are dotted with penultimate layer of neural network to get the final prediction. OSA is evaluated with different neural networks on ICLR 2017—2018 datasets, experimental results verify that OSA is of high efficiency and achieves outstanding performance on two datasets.

Key words: peer review, sentiment analysis, pu-learning, data mining

林原, 王凯巧, 杨亮, 林鸿飞, 任璐, 丁堃. 基于pu-learning的同行评议文本情感分析[J]. 计算机工程与应用, 2023, 59(3): 143-149.

LIN Yuan, WANG Kaiqiao, YANG Liang, LIN Hongfei, REN Lu, DING Kun. Sentiment Analysis of Peer Review Texts Based on Pu-Learning[J]. Computer Engineering and Applications, 2023, 59(3): 143-149.

参考文献

[1] BORNMANN L，DANIEL H D.Reliability of reviewers’ ratings when using public peer review：a case study[J].Learned Publishing，2010，23（2）：124-131.
[2] 徐志英.科学文章同行评议研究进展[J].中国科技期刊研究，2014，25（11）：1355-1359.
XU Z Y.Research progress of peer review of scientific articles[J].Chinese Journal of Scientific and Technical Periodicals，2014，25（11）：1355-1359.
[3] 秦成磊，章成志.大数据环境下同行评议面临的问题与对策[J].情报理论与实践，2021，44（4）：99-112.
QIN C L，ZHANG C Z.Problems and responses strategy of peer review in the context of big data[J].Information Studies：Theory & Application，2021，44（4）：99-112.
[4] TABOADA M，BROOKE J，TOFILOSKI M，et al.Lexicon-based methods for sentiment analysis[J].Computational Linguistics，2011，37（2）：267-307.
[5] TURNEY P.Thumbs up or thumbs down？Semantic orientation applied to unsupervised classification of reviews[C]//Proceeding of the 40th Annual Meeting of the Association for Computational Linguistics，2002：417-424.
[6] BLITZER J，DREDZE M，PEREIRA F.Biographies，bollywood，boom-boxes and blenders：domain adaptation for sentiment classification[C]//Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics，2007：440-447.
[7] GO A，BHAYANI R，LEI H.Twitter sentiment classification using distant supervision[R/OL].CS224N project report，Stanford，2009.https：//cs.stanford.edu/people/alecmgo/papers/TwitterDistantSupervision09.pdf.
[8] DERIU J，LUCCHI A，LUCA V D，et al.Leveraging large amounts of weakly supervised data for multi-language sentiment classification[C]//Proceedings of the 26th International Conference on World Wide Web，2017：1045-1052.
[9] KIM Y.Convolutional neural networks for sentence classification[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing，Doha，Qatarjc，2014：1746-1751.
[10] REN Y，ZHANG Y，ZHANG M，et al.Context-sensitive twitter sentimentclassification using neural network[C]//Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence，2016：215-221.
[11] DAVE K，LAWRENCE S，PENNOCK D M，et al.Mining the peanutgallery：opinion extraction and semantic classification of product reviews[C]//Proceedings of International Conference on World Wide Web，2003：519-528.
[12] PANG B，LEE L，VAITHYANATHAN S.Thumbs up? Sentiment classification using machine learning techniques[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing，2002：79-86.
[13] GODBOLE N，SRINIVASAIAH M，SKIENA S.Large-scale sentiment analysis for news and blogs[C]//Proceedings of the First International Conference on Weblogs and Social Media，2007：1-4.
[14] TESSER A，MARTIN L.Reviewing empirical submissions to journals[M]//STERNBERG R.Reviewing scientific works in psychology，2006：3-30.
[15] LOVEJOY T I，REVENSON T A，FRANCE C R.Reviewing manuscripts for peer-review journals：a primer for novice and seasoned reviewers[J].Annals of Behavioral Medicine，2011，42（1）：1-13.
[16] KABOUTARI A，BAGHERZADEH J，KHERADMAND F.An evaluation of two-step techniques for positive-unlabeled learning in text classification[J].International Journal of Computer Applications Technology and Research，2014，3（9）：592-594.
[17] DEVLIN J，CHANG M W，LEE K，et al.BERT：pre-training of deep bidirectional transformers for language understanding[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics：Human Language Technologies，2019：4171-4186.
[18] KABOUTARI A，BAGHERZADEH J，KHERADMAND F.An evaluation of two-step techniques for positive-unlabeled learning in text classification[J].International Journal of Computer Applications Technology and Research，2014，3（9）：592-594.
[19] GRAVES A.Supervised sequence labelling with recurrent neural networks[J].Studies in Computational Intelligence，2012，385：13-148.