融合多语言知识的慕课评论隐式方面情感分析

doi:10.3778/j.issn.1002-8331.2405-0163

摘要/Abstract

摘要： 慕课完成率不高的问题严重制约着其高质量发展，慕课评论中隐喻、客观事实描述、讽刺、反问等表达中蕴含的隐式情感更为真实地表达了用户的学习体验，对信息进行分析、利用，从而挖掘出学生关于课程的反馈信息，并做出相应的改善，有助于提升学生满意度以提高慕课完成率。为此，提出一种融合多语言知识的慕课隐式方面情感分析模型来获得更为精准的隐式情感信息。针对前两种表达中缺乏明显情感倾向的特点，引入多重图神经网络来融合词性、语义、句法和义原等多语言知识，充分利用其中的关联关系来挖掘评论中隐含的情感信息。同时，对于后两种表达方式中的情感词与文本真实情感极性不符的问题，构建多层级注意力机制来获取整体语义粗粒度、方面词细粒度中的情感信息。在构建的MOOC数据集上测试模型，准确率和F1指数分别达到90.2%和93.8%，同时在SMP2019-ECISA数据集上的对比实验表明，所提模型的准确率与KC-ISA-BERT等模型相比提升了1.7个百分点。

关键词: 隐式情感分析, 方面情感分析, 图神经网络, 多级注意力机制, 慕课

Abstract: The issue of low completion rates in MOOCs severely restricts their high-quality development. The implicit emotions contained in expressions such as metaphors, objective factual descriptions, sarcasm, and rhetorical questions in MOOC comments more genuinely reflect users’ learning experiences. Analyzing and utilizing this information to uncover student feedback on the courses and make corresponding improvements can help enhance MOOC completion rates. To this end, the paper proposes a MOOC implicit aspect sentiment analysis model integrating multilingual knowledge to obtain more accurate implicit sentiment information. To address the lack of clear emotional tendencies in the first two expressions, a multi-graph neural network is introduced to combine multi-language knowledge such as part of speech, semantics, syntax, and semantic primitives, fully utilizing the associated relationships to uncover implicit emotional information in comments. Meanwhile, to address the issue of emotional words not matching the true sentiment polarity in the last two expression methods, a multi-level attention mechanism is constructed to capture emotional information at both the coarse-grained level of overall semantics and the fine-grained level of aspect words. Testing the model on the MOOC dataset for paper construction achieves accuracy and F1 scores of 90.2% and 93.8%, respectively. Comparative experiments on SMP2019-ECISA dataset reveals an improved accuracy of the proposed model by 1.7 percentage points compared to models like KC-ISA-BERT.

Key words: implicit sentiment analysis, aspect-based sentiment analysis, graph neural network, multi-polarity attention, massive open online course(MOOC)

陈怀博, 张会兵, 首照宇, 潘芳. 融合多语言知识的慕课评论隐式方面情感分析[J]. 计算机工程与应用, 2025, 61(5): 104-112.

CHEN Huaibo, ZHANG Huibing, SHOU Zhaoyu, PAN Fang. Integrating Multilingual Knowledge for Implicit Aspect-Based Sentiment Analysis of MOOC Reviews[J]. Computer Engineering and Applications, 2025, 61(5): 104-112.

参考文献

[1] 张新香, 段燕红. 基于学习者在线评论文本的MOOC质量评判——以“中国大学MOOC”网的在线评论文本为例[J]. 现代教育技, 2020, 30（9）: 56-63.
ZHANG X X, DUAN Y H. MOOC quality evaluation based on learners online review texts—taking the online review texts from the “China university MOOC” website as examples[J]. Modern Educational Technology, 2020, 30（9）: 56-63.
[2] LIAO J, WANG M, CHEN X, et al. Dynamic commonsense knowledge fused method for Chinese implicit sentiment analysis[J]. Information Processing & Management, 2022, 59（3）: 102934-102949.
[3] LIAO J, WANG S, LI D. Identification of fact- implied implicit sentiment based on multi-level semantic fused representation[J]. Knowledge- Based Systems, 2019, 165: 197-207.
[4] WANG J, LI X, HE J, et al. Enhancing implicit sentiment learning via the incorporation of part-of-speech for aspect-based sentiment analysis[C]//Proceedings of the China National Conference on Chinese Computational Linguistics, 2023: 382-399.
[5] YANG S, XING L, LI Y, et al. Implicit sentiment analysis based on graph attention neural network[J]. Engineering Reports, 2022, 4（1）: 12452-12468.
[6] DENG C, LAI G, DENG H. Improving word vector model with part-of‐speech and dependency grammar information[J]. CAAI Transactions on Intelligence Technology, 2020, 5（4）: 276-282.
[7] SUHROB E, VASILA K. Parts of speech and sentence structure in English grammar[J]. Galaxy International Interdisciplinary Research Journal, 2022, 10（7）: 156-160.
[8] 杨春霞, 韩煜, 陈启岗, 等. 基于BERT与注意力机制的方面级隐式情感分析模型[J]. 南京信息工程大学学报（自然科学版）, 2023, 15（5）: 551-560.
YANG CX, HAN Y, CHEN Q G, et al. Aspect-based implicit sentiment analysis model based on BERT and attention mechanism[J]. Journal of Nanjing University of Information Science & Technology（Nature Sciences）, 2023, 15（5）: 551-560.
[9] BERARAFA H, BENKHALIFA M, AKHLOUFI M. WordNet semantic relations based enhancement of KNN model for implicit aspect identification in sentiment analysis[J]. International Journal of Computational Intelligence Systems, 2023, 16（1）: 3-17.
[10] QIAN Y, WANG J, LI D, et al. Interactive capsule network for implicit sentiment analysis[J]. Applied Intelligence, 2023, 53（3）: 3109-3123.
[11] ZHANG L, LIU B. Identifying noun product features that imply opinions[C]//Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, 2011: 575-580.
[12] LIU B. Sentiment analysis and opinion mining[M]. Cham: Springer , 2012.
[13] CHOI Y, WIEBE J. +/-effectwordnet: senselevel lexicon acquisition for opinion inference[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing, 2014: 1181-1191.
[14] 谈光璞, 朱广丽, 韦斯羽. 基于情感特征增强的中文隐式情感分类模型[J]. 计算机工程与应用, 2024, 60（3）: 196-204.
TAN G P , ZHU G L , WEI S Y. Implicit sentiment classification model based on enhancement of sentiment features oriented to Chinese text[J]. Computer Engineering and Applications, 2024, 60（3）: 196-204.
[15] SU C, LI J, PENG Y, et al. Implicit mood computing via LSTM and semantic mapping[J] Soft Computing, 2020, 24: 15795-15809.
[16] WEI J, LIAO J, YANG Z, et al. BiLSTM with multi-polarity orthogonal attention for implicit sentiment analysis[J]. Neurocomputing, 2020, 383: 65-173.
[17] YIN W, SHANG L. ContextBERT: enhanced implicit sentiment analysis using implicit- sentiment-query attention[C]//Proceedings of the 2022 International Joint Conference on Neural Networks, 2022: 1-8.
[18] GREENE S, RESNIK P. More than words: syntactic packaging and implicit sentiment[C]//Proceedings of the Human Language Technologies: the 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, 2009: 503-511.
[19] FU L, LIU S. A syntax-based BSGCN model for Chinese implicit sentiment analysis with multi-classification[C]//Proceedings of the 2022 IEEE 16th International Conference on Application of Information and Communication Technologies, 2022: 1-7.
[20] LI R, CHEN H, FENG F, et al. DualGCN: exploring syntactic and semantic information for aspect-based sentiment analysis[J]. IEEE Transactions on Neural Networks and Learning Systems, 2024, 35(6): 7642-7656.
[21] 田继帅, 艾芳菊. 基于增强句法信息与多特征图卷积融合的方面级情感分析[J/OL]. 计算机科学与探索: 1-15 [2024-07-25]. http://kns.cnki.net/kcms/detail/11.5602.tp.20240724.115
3.006.html.
TIAN J S, AI F J. Aspect-level sentiment analysis based on enhanced syntactic information and multi-feature graph convolutional fusion[J/OL]. Journal of Frontiers of Computer Science and Technology: 1-15[2024-07-25]. http://kns.cnki.net/kcms/detail/11.5602.tp.20240724.1153.006.html.
[22] XU M, WANG D, FENG S, et al. KC-ISA: an implicit sentiment analysis model combining knowledge enhancement and context features[C]//Proceedings of the 29th International Conference on Computational Linguistics, 2022: 6906-6915.
[23] AHMED M, PAN S, SU J, et al. BERT-ASC: implicit aspect representation learning through auxiliary-sentence construction for sentiment analysis[J]. arXiv:2203.11702, 2022.
[24] WANG H, HOU M. Quantum-like implicit sentiment analysis with sememes knowledge[J]. Expert Systems with Applications, 2023, 232: 120720-120729.
[25] YING Z, YOU J, MORRIS C, et al. Hierarchical graph representation learning with differentiable pooling[J]. arXiv:1806.08804, 2018.
[26] DONG Z, DONG Q. HowNet a hybrid language and knowledge resource[C]//Proceedings of the International Conference on Natural Language Processing and Knowledge Engineering, 2003: 820-824.
[27] RONG Y, HUANG W, XU T, et al. Dropedge: towards deep graph convolutional networks on node classification[J]. arXiv:1907.10903, 2019.
[28] ZHOU P, SHI W, TIAN J, et al. Attention-based bidirectional long short-term memory networks for relation classification[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, 2016: 207-212.
[29] JOHNSON R, ZHANG T. Deep pyramid convolutional neural networks for text categorization[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, 2017: 562-570.