簡易檢索 / 詳目顯示

回結果列表
研究生： 潘楚達
Putra, Cendra Devayana
論文名稱： 半監督式仇恨言論偵測方法
Semi Meta Supervised for Hate Speech Detection
指導教授： 王惠嘉
Wang, Hei-Chia
學位類別： 碩士
Master
系所名稱： 管理學院 - 資訊管理研究所
Institute of Information Management
論文出版年： 2022
畢業學年度： 110
語文別： 英文
論文頁數： 59
中文關鍵詞： 半監督 、單任務學習 、共享知識 、仇恨言論
外文關鍵詞： Semi-Supervised, Single-Task Learning, Shared Knowledge, Hate Speech
ORCID： 0000-0002-5692-9762
ResearchGate： https://www.researchgate.net/profile/Cendra-Devayana-Putra
相關次數： 點閱：82  下載：9 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 在社交媒體上，仇恨言論屢見不鮮。這種現象具有多種生理和心理後果。因此，必須採取適當的方法來防止仇恨言論。防止仇恨言論的一種方法是通過深度學習過程。然而，深度學習方法需要一個大數據集來生成一個好的模型。事實上，仇恨言論的數據集是有缺陷且有偏差的。我們提出了 Semi BERT-SP 來解決這個問題。我們利用這種偏見通過分享相關知識來改進深度學習模型。 Semi BERT-SP 模型包含多項改進，包括 BERT 準確度的提高、共享學習準確度的提高以及共享數據集準確度的提高。此外，我們將我們的最大準確度與其他五個深度學習模型進行了比較。 Semi BERT-SP 在戴維森數據集上實現了 97% 的準確率和 93% 的 F1 分數，在吉爾伯特數據集上實現了 71% 的準確率和 70% 的 F1 分數，在 DataTurks 上實現了 93% 的準確率和 92% 的 F1 分數，62% 的準確率和 61% F1在 Kumar 上得分，在 Bhattacharya 上獲得 92% 的準確率和 61% 的 F1 得分。作為這項研究的結果，我們的模型幾乎優於所有基準。

    On social media, hate speech is a common occurrence. This phenomenon has a variety of physical and psychological consequences. As a result, a proper method to prevent hate speech is mandatory. One method of preventing hate speech is through a process of deep learning. However, the deep learning approach requires a large dataset to produce a good model. Indeed, the dataset on hate speech is deficient and skewed. We leverage this bias to improve deep learning models by sharing pertinent knowledge. Our proposed model incorporates several improvements, including an increase in BERT accuracy, an increase in shared learning accuracy, and an increase in shared dataset accuracy. Additionally, we compared our maximum accuracy to five other deep learning models. As a result of this research, our model outperforms almost all benchmarks.

    Abstracts................................................................................................III 摘要.........................................................................................................IV Acknowledgments................................................................................V Table of Contents..................................................................................VI List of Tables...........................................................................................VIII List of Figures...........................................................................................IX CHAPTER 1 INTRODUCTION...................................................................1 1.1 Background and Motivation.........................................................1 1.2 Objectives.......................................................................................3 1.3 Scope and Limitation.....................................................................4 1.4 Research Process............................................................................4 1.5 Research Overview........................................................................6 CHAPTER 2 LITERATURE REVIEW.............................................................7 2.1 Hate Speech Definition..................................................................7 2.2 Hate Speech Detection.................................................................8 2.3 Deep Bidirectional Transformer Encoder.....................................9 2.4 Semi Meta Supervised Learning..................................................10 2.5 Summary of Related Works..........................................................11 CHAPTER 3 RESEARCH METHODS........................................................13 3.1 Theoretical Framework................................................................13 3.2 Data Preprocessing......................................................................15 3.3 Single-Task Learning......................................................................17 3.3.1 Deep Bidirectional Transformer Encoder...................18 3.3.2 Convolution Neural Network.......................................25 3.4 Shared-Private Learning...............................................................27 3.4.1 Shared Layer Training...................................................28 3.4.2 Private Layer Training...................................................29 3.5 Semi Supervised............................................................................29 3.5.1 Pseudo label Selection................................................30 3.5.2 Hard & Soft Knowledge Sharing..................................31 3.6 Summary of Proposed Module....................................................32 CHAPTER 4 EXPERIMENTS AND ANALYSIS...........................................33 4.1 Experiment Environment Setup....................................................33 4.2 Experiment Method.......................................................................34 4.2.1 Hate Speech Datasets..................................................34 4.2.2 Experiment Design..........................................................38 4.2.3 Evaluation Matrix............................................................38 4.2.4 Validation Methods........................................................40 4.3 Parameter Settings.........................................................................42 4.3.1 Single-Task Learning........................................................42 4.3.2 Share-Private Layer.........................................................44 4.3.3 Semi-Supervised BERT with Shared-Private Layer.........45 4.4 Overall Benchmark Comparisons...................................................48 CHAPTER 5 CONCLUSIONS.....................................................................51 5.1 Contribution and Discussion...........................................................51 5.2 Future Research Work.....................................................................52 REFERENCES.............................................................................................54

    Abas, A. R., Elhenawy, I., Zidan, M., & Othman, M. (2022). BERT-CNN: A Deep Learning Model for Detecting Emotions from Text. Computers, Materials and Continua, 71(2), 2943–2961. https://doi.org/10.32604/cmc.2022.021671
    Al-Laith, A., Shahbaz, M., Alaskar, H. F., & Rehmat, A. (2021). AraSenCorpus: A Semi-Supervised Approach for Sentiment Annotation of a Large Arabic Text Corpus. Applied Sciences (Switzerland), 11(5). https://doi.org/10.3390/app11052434
    Albert Gonzalez Farran. (2021). Discrimination and Hate Speech Fuel Violence in Sudan. United Nations Report, March. https://reliefweb.int/sites/reliefweb.int/files/resources/b476a0_7aeeadba08ff4732806c833b72879cbc.pdf
    Ayo, F. E. F. E., Folorunso, O., Ibharalu, F. T. F. T. F. T., Osinuga, I. A. I. A., & Abayomi-Alli, A. (2021). A Probabilistic Clustering Model for Hate Speech Classification in Twitter. Expert Systems with Applications, 173(February 2020), 114762. https://doi.org/10.1016/j.eswa.2021.114762
    Ba, J. L., Kiros, J. R., & Hinton, G. E. (2016). Layer Normalization. ArXiv. http://arxiv.org/abs/1607.06450
    Baydogan, C., & Alatas, B. (2021). Metaheuristic Ant Lion and Moth Flame Optimization-Based Novel Approach for Automatic Detection of Hate Speech in Online Social Networks. IEEE Access, 9, 110047–110062. https://doi.org/10.1109/ACCESS.2021.3102277
    Bhatt, S., Patwa, F., & Sandhu, R. (2017). Natural Language Processing (Almost) from Scratch. Journal of Machine Learning Research, 2017-Janua, 328–338. https://doi.org/10.1109/CIC.2017.00050
    Bhattacharya, S., Singh, S., Kumar, R., Bansal, A., Bhagat, A., Dawer, Y., Lahiri, B., & Ojha, A. K. (2020). Developing a Multilingual Annotated Corpus of Misogyny and Aggression. ArXiv. http://arxiv.org/abs/2003.07428
    Cao, R., Lee, R. K.-W. R. K. W., & Hoang, T.-A. T. A. (2020). DeepHate: Hate Speech Detection via Multi-Faceted Text Representations. WebSci 2020 - Proceedings of the 12th ACM Conference on Web Science, 11–20. https://doi.org/10.1145/3394231.3397890
    Capistrano, J. L. C., Suarez, J. J. P., & Naval, P. C. (2019). SALSA:Detection of Cybertrolls using Sentiment, Aggression, Lexical and Syntactic Analysis of Tweets. ACM International Conference Proceeding Series. https://doi.org/10.1145/3326467.3326471
    Chakrabarty, T., Gupta, K., & Muresan, S. (2019). Pay “Attention” to your Context when Classifying Abusive Language. 2017, 70–79. https://doi.org/10.18653/v1/w19-3508
    Chinchor, N. (1992). MUC-4 evaluation metrics. Association for Computing Machinery, 22. https://doi.org/10.3115/1072064.1072067
    Clement, J. (2020, August). Facebook MAU worldwide 2020. Statista.Com.
    Cruz, R. M. O. O., de Sousa, W. V., & Cavalcanti, G. D. C. C. (2022). Selecting and Combining Complementary Feature Representations and Classifiers for Hate Speech Detection. Online Social Networks and Media, 28(March 2021), 100194. https://doi.org/10.1016/j.osnem.2021.100194
    DataTurks. (2018). Tweets Dataset for Detection of Cyber-Trolls. Kaggle. https://www.kaggle.com/dataturks/dataset-for-detection-of-cybertrolls
    Davidson, T., Warmsley, D., Macy, M., & Weber, I. (2017). Automated Hate Speech Detection and the Problem of Offensive Language. Proceedings of the 11th International Conference on Web and Social Media, ICWSM 2017, 512–515. www.facebook.
    de Gibert, O., Perez, N., García-Pablos, A., & Cuadros, M. (2019). Hate Speech Dataset from a White Supremacy Forum. 11–20. https://doi.org/10.18653/v1/w18-5102
    DeMarsico, D., Bounoua, N., Miglin, R., & Sadeh, N. (2021). Aggression in the Digital Era: Assessing the Validity of the Cyber Motivations for Aggression and Deviance Scale. Assessment. https://doi.org/10.1177/1073191121990088
    Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2019). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference, 1(Mlm), 4171–4186.
    Dijkstra, K. (1987). Self-taught Learning: Transfer Learning from Unlabeled Data. The Journal of the Acoustical Society of America, 82(2), 726–726. https://doi.org/10.1121/1.395361
    Donahue, J., Hoffman, J., Rodner, E., Saenko, K., & Darrell, T. (2013). Semi-Supervised Domain Adaptation with Instance Constraints. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 668–675. https://doi.org/10.1109/CVPR.2013.92
    Facebook Company. (2022). Community Standarts. Facebook. https://transparency.fb.com/en-gb/policies/community-standards/
    Fortuna, P., & Nunes, S. (2018). A Survey on Automatic Detection of Hate Speech in Text. ACM Computing Surveys, 51(4). https://doi.org/10.1145/3232676
    Founta, A. M. A.-M. A. M., Chatzakou, D., Kourtellis, N., Blackburn, J., Vakali, A., & Leontiadis, I. (2019). A Unified Deep Learning Architecture for Abuse Detection. WebSci 2019 - Proceedings of the 11th ACM Conference on Web Science, 105–114. https://doi.org/10.1145/3292522.3326028
    Francesca Farmer. (2021). Through These Walls. United Nations Report, 1–34.
    He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep Residual Learning for Image Recognition. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2016-Decem, 770–778. https://doi.org/10.1109/CVPR.2016.90
    Hinton, G., Vinyals, O., & Dean, J. (2015). Distilling the Knowledge in a Neural Network. ArXiv, 1–9. http://arxiv.org/abs/1503.02531
    Howard, J. W. (2019). Free Speech and Hate Speech. Annual Review of Political Science, 22, 93–109. https://doi.org/10.1146/annurev-polisci-051517-012343
    Kapil, P., Ekbal, A., & Das, D. (2020). Investigating Deep Learning Approaches for Hate Speech Detection in Social Media. ArXiv, 1–12.
    Kapil, Prashant, & Ekbal, A. (2020). A Deep Neural Network Based Multi-Task Learning Approach To Hate Speech Detection. Knowledge-Based Systems, 210, 106458. https://doi.org/10.1016/j.knosys.2020.106458
    Khan, M. U. S. M. U. S., Abbas, A., Rehman, A., & Nawaz, R. (2021). HateClassify: A Service Framework for Hate Speech Identification on Social Media. IEEE Internet Computing, 25(1), 40–49. https://doi.org/10.1109/MIC.2020.3037034
    Khandelwal, A., & Kumar, N. (2020). A Unified System For Aggression Identification In English Code-Mixed And Uni-Lingual Texts. ACM International Conference Proceeding Series, 55–64. https://doi.org/10.1145/3371158.3371165
    Kiritchenko, S., Zhu, X., & Mohammad, S. M. (2014). Sentiment Analysis of Short Informal Texts. Journal of Artificial Intelligence Research, 50, 723–762. https://doi.org/10.1613/jair.4272
    Kumar, R., Bhanodai, G., Pamula, R., & Chennuru, M. R. (2018). TRAC-1 Shared Task on Aggression Identification: IIT(ISM)@COLING’18. First Workshop on Trolling, Aggression and Cyberbullying, 58. https://github.com/zeerakw/hatespeech
    Kumar, R., Lahiri, B., & Ojha, A. K. (2021). Aggressive and Offensive Language Identification in Hindi, Bangla, and English: A Comparative Study. SN Computer Science, 2(1), 1–20.
    Kumar, V., Pasari, S., Patil, V. P., & Seniaray, S. (2020). Machine Learning based Language Modelling of Code Switched Data. Proceedings of the International Conference on Electronics and Sustainable Communication Systems, ICESC 2020, 552–557. https://doi.org/10.1109/ICESC48915.2020.9155695
    Liu, H., Burnap, P., Alorainy, W., & Williams, M. L. M. L. (2019). A Fuzzy Approach to Text Classification With Two-Stage Training for Ambiguous Instances. IEEE Transactions On Computational Social Systems, 6(2), 227–240. https://doi.org/10.1109/TCSS.2019.2892037
    Liu, P., Qiu, X., & Huang, X. (2017). Adversarial Multi-task Learning for Text Classification. ACL 2017 - 55th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers), 1, 1–10. https://doi.org/10.18653/v1/P17-1001
    MacAvaney, S., Yao, H.-R. H.-R. R., Yang, E., Russell, K., Goharian, N., & Frieder, O. (2019). Hate Speech Detection: Challenges and Solutions. Plos One, 14(8), 1–16. https://doi.org/10.1371/journal.pone.0221152
    Madukwe, K. J. K. J., & Gao, X. (2019). The Thin Line Between Hate and Profanity. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 11919 LNAI, 344–356. https://doi.org/10.1007/978-3-030-35288-2_28
    Malmasi, S., & Zampieri, M. (2017). Detecting Hate Speech in Social Media. International Conference Recent Advances in Natural Language Processing, RANLP, 2017-Septe, 467–472. https://doi.org/10.26615/978-954-452-049-6-062
    Modha, S., Majumder, P., Mandl, T., & Mandalia, C. (2020). Detecting and Visualizing Hate Speech in Social Media: A Cyber Watchdog for Surveillance. Expert Systems with Applications, 161, 113725. https://doi.org/10.1016/j.eswa.2020.113725
    Mossie, Z., & Wang, J.-H. H. J.-H. (2020). Vulnerable community identification using hate speech detection on social media. Information Processing & Management, 57(3), 102087. https://doi.org/10.1016/j.ipm.2019.102087
    Mozafari, M., Farahbakhsh, R., & Crespi, N. (2020a). A BERT-Based Transfer Learning Approach for Hate Speech Detection in Online Social Media. Studies in Computational Intelligence, 881 SCI, 928–940. https://doi.org/10.1007/978-3-030-36687-2_77
    Mozafari, M., Farahbakhsh, R., & Crespi, N. N. (2020b). Hate Speech Detection and Racial Bias Mitigation in Social Media Based on BERT Model. Plos One, 15(8), e0237861. https://doi.org/10.1371/journal.pone.0237861
    Naseem, U., Razzak, I., & Eklund, P. W. P. W. (2020). A Survey of Pre-Processing Techniques to Improve Short-Text Quality: A Case Study on Hate Speech Detection on Twitter. Multimedia Tools and Applications. https://doi.org/10.1007/s11042-020-10082-6
    Naseem, U., Razzak, I., & Hameed, I. A. (2019). Deep Context-Aware Embedding for Abusive and Hate Speech detection on Twitter. Journal of Chemical Information and Modeling, 53(9), 1689–1699.
    Oriola, O., & Kotze, E. (2020). Evaluating Machine Learning Techniques for Detecting Offensive and Hate Speech in South African Tweets. IEEE Access, 8, 21496–21509. https://doi.org/10.1109/ACCESS.2020.2968173
    Pamungkas, E. W. E. W., Basile, V., & Patti, V. (2020). Misogyny Detection in Twitter: a Multilingual and Cross-Domain Study. Information Processing & Management, 57(6). https://doi.org/10.1016/j.ipm.2020.102360
    Paul, S., & Saha, S. (2020). Cyberbert: BERT for Cyberbullying Identification. Multimedia Systems, 1, 3. https://doi.org/10.1007/s00530-020-00710-4
    Pham, H., Dai, Z., Xie, Q., Luong, M.-T., & Le, Q. V. (2021). Meta Pseudo Labels. Google Research. https://arxiv.org/abs/2003.10580v4
    Rahali, A., Akhloufi, M. A., Therien-Daniel, A. M., & Brassard-Gourdeau, E. (2021). Automatic Misogyny Detection in Social Media Platforms using Attention-based Bidirectional-LSTM*. Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, 2706–2711. https://doi.org/10.1109/SMC52423.2021.9659158
    Refaeilzadeh, P., Tang, L., & Liu, H. (2009). Cross-Validation. In Encyclopedia of Database Systems (pp. 532–538). Springer, Boston, MA. https://doi.org/10.1007/978-0-387-39940-9_565
    Rusu, A. A., Rabinowitz, N. C., Desjardins, G., Soyer, H., Kirkpatrick, J., Kavukcuoglu, K., Pascanu, R., & Hadsell, R. (2016). Progressive Neural Networks. http://arxiv.org/abs/1606.04671
    Saha, K., Chandrasekharan, E., & De Choudhury, M. (2019). Prevalence and psychological effects of hateful speech in online college communities. WebSci 2019 - Proceedings of the 11th ACM Conference on Web Science, 10, 255–264. https://doi.org/10.1145/3292522.3326032
    Samghabadi, N. S., Patwa, P., Pykl, S., Mukherjee, P., Das, A., & Solorio, T. (2020). Aggression and Misogyny Detection using BERT: A Multi-Task Approach. Proceedings of the Second Workshop on Trolling, Aggression and Cyberbullying, May, 126–131. https://www.theverge.com/interface/2019/
    Shruthi, P., Kumar, A. K. M., Anil Kumar, K. M. M., & Kumar, A. K. M. (2020). Novel Approach for Generating Hybrid Features Set to Effectively Identify Hate Speech. Inteligencia Artificial, 23(66), 97–111. https://doi.org/10.4114/intartif.vol23iss66pp97-111
    Shuang, K., Xu, M., Zhang, W., & Zhang, Z. (2019). Adversarial Multi-task Label Embedding for Text Classification. ACM International Conference Proceeding Series, 45–50. https://doi.org/10.1145/3372422.3372433
    Spertus, E. (1997). Smokey: automatic recognition of hostile messages. Innovative Applications of Artificial Intelligence - Conference Proceedings, 1058–1065.
    Sulaiman, N. S., Shafiq, A., & Ahmad, H. (2021). Comparison of Operating System Performance Between Windows 10 and Linux Mint. 2(1), 92–102.
    Twitter. (2017). Hateful Conduct. In Https://Help.Twitter.Com/En/Rules-and-Policies/Hateful-Conductpolicy.
    Twitter, & Clement, J. (2020). Monthly Active Users Worldwide. In Statista.
    United Nations. (2019). United Nations Strategy and Plan of Action on Hate Speech. United Nations Report, May, 1–5.
    van Aken, B., Risch, J., Krestel, R., & Löser, A. (2018). Challenges for Toxic Comment Classification: An In-Depth Error Analysis. 2nd Workshop on Abusive Language Online - Proceedings of the Workshop, Co-Located with EMNLP 2018, 33–42. https://doi.org/10.18653/v1/w18-5105
    Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., & Polosukhin, I. (2017). Attention Is All You Need. Advances in Neural Information Processing Systems, 2017-Decem, 5999–6009. https://arxiv.org/abs/1706.03762v5
    Wu, H., & Gu, X. (2015). Max-Pooling Dropout for Regularization of Convolutional Neural Networks. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 9489, 46–54. https://doi.org/10.1007/978-3-319-26532-2_6
    Wu, Yanzhao, Liu, L., Pu, C., Cao, W., Sahin, S., Wei, W., & Zhang, Q. (2019). A Comparative Measurement Study of Deep Learning as a Service Framework. IEEE Transactions on Services Computing, Dl. https://doi.org/10.1109/TSC.2019.2928551
    Wu, Yonghui, Schuster, M., Chen, Z., Le, Q. V, Norouzi, M., Macherey, W., Krikun, M., Cao, Y., Gao, Q., Macherey, K., Klingner, J., Shah, A., Johnson, M., Liu, X., Kaiser, Ł., Gouws, S., Kato, Y., Kudo, T., Kazawa, H., … Dean, J. (2016). Google’s Neural Machine Translation System: Bridging the Gap between Human and Machine Translation. http://arxiv.org/abs/1609.08144
    Wullach, T., Adler, A., & Minkov, E. (2021). Towards Hate Speech Detection at Large via Deep Generative Modeling. IEEE Internet Computing, 25(2), 48–57. https://doi.org/10.1109/MIC.2020.3033161

    下載圖示 校內：2024-06-28公開
    校外：2024-06-28公開
    QR CODE