1. NCCU Academic Hub
  2. 學術產出
  3. 資訊學院
  4. 學位論文
  5. Item

學術產出-學位論文

文章檢視/開啟

書目匯出

Google ScholarTM

政大圖書館

引文資訊

TAIR相關學術產出

題名 以深度遞歸神經網路實施多重任務學習偵測假新聞
Deep Recurrent Neural Networks with Multi-Task Learning for Fake News Detection
作者 劉永鈞
Liou, Yung-Jiun
貢獻者 胡毓忠
Hu, Yuh-Jong
劉永鈞
Liou, Yung-Jiun
關鍵詞 社交媒體
假新聞
錯誤資訊
多重任務學習
假新聞資料集
遞歸神經網路
傳統深度學習
Social Media
Fake News
Misinformation
Muti-Task Learning
PHEME
Fake News Dataset
Recurrent Neural Network
GRU
Traditional Deep Learning
日期 2020
上傳時間 2-三月-2020 11:38:01 (UTC+8)
摘要 偵測假新聞是一項十分艱鉅的任務，包含偵測假新聞(Rumour Detection)、假新聞追蹤(Rumour Tracking)及立場分類(Stance Classification)，從這些方法最終對假新聞作驗證(Rumour Verification)。欲做到辨識新聞的驗證以使讀者能閱讀到正確的新聞及資訊，本研究希望探索以多重任務學習(Multi-Task Learning, MTL)用於處理數量龐大的假新聞資料上，並比較與傳統深度學習的差異，達到自動辨識及判別假新聞的目的。
本研究使用RumourEval、PHEME兩種假新聞資料集來進行深度遞歸神經網路(Recurrent Neural Network, RNN)中的GRU（Gate Recurrent Unit）演算法實作，並進行多重任務學習的訓練，對假新聞進行分類，進而找出處理識別假新聞的最佳參數。最後透過各種模擬實驗來比較改良過後的深度學習演算法(即GRU)與傳統深度學習的差異，並依據實驗結果進行量化與質化的分析。
Detecting fake news is a very difficult task, including Rumour Detection,Rumour Tracking and Stance Classification, and finally leading to Rumour Verification. To identify the authenticity of news so that readers can read the correct news and information, this research hopes to explore the use of Multi-Task Learning technology for processing a large number of fake news datasets and compare it with traditional deep learning, to achieve the purpose of automatically identifying and distinguishing fake news.
This research uses two fake news datasets, RumourEval and PHEME,to implement the GRU (Gated Recurrent Unit) algorithm of the Recurrent Neural Network (RNN), and trains for multiple tasks to perform fake news classification to find the best parameters for handling fake news. Finally, through various simulation experiments, the differences between the improved and traditional deep learning algorithm will be compared, and quantitative and qualitative analysis is performed based on the experimental results
參考文獻 [1] P. N. Howard, G. Bolsover, B. Kollanyi, S. Bradshaw, and L.-M. Neudert,
\\Junk news and bots during the us election: What were michigan voters
sharing over twitter," CompProp, OII, Data Memo, 2017.
[2] E. Kochkina, M. Liakata, and A. Zubiaga, \\All-in-one: Multi-task learning
for rumour verification," arXiv preprint arXiv:1806.03713, 2018.
[3] A. Zubiaga, A. Aker, K. Bontcheva, M. Liakata, and R. Procter, \\Detection and resolution of rumours in social media: A survey," ACM Computing
Surveys (CSUR), vol. 51, no. 2, p. 32, 2018.
[4] Z. Zhao, P. Resnick, and Q. Mei, \\Enquiring minds: Early detection of rumors
in social media from enquiry posts," in Proceedings of the 24th International
Conference on World Wide Web. International World Wide Web Conferences
Steering Committee, 2015, pp. 1395{1405.
[5] L. Derczynski, K. Bontcheva, M. Liakata, R. Procter, G. W. S. Hoi, and
A. Zubiaga, \\Semeval-2017 task 8: Rumoureval: Determining rumour veracity and support for rumours," arXiv preprint arXiv:1704.05972, 2017.
[6] M. Mendoza, B. Poblete, and C. Castillo, \\Twitter under crisis: Can we trust
what we rt?" in Proceedings of the first workshop on social media analytics.
ACM, 2010, pp. 71{79.
[7] R. Procter, F. Vis, and A. Voss, \\Reading the riots on twitter: methodological innovation for the analysis of big data," International journal of social
research methodology, vol. 16, no. 3, pp. 197{214, 2013.
[8] M. Lukasik, P. Srijith, D. Vu, K. Bontcheva, A. Zubiaga, and T. Cohn,
\\Hawkes processes for continuous time sequence classification: an application to rumour stance classification in twitter," in Proceedings of the 54th
Annual Meeting of the Association for Computational Linguistics (Volume 2:
Short Papers), vol. 2, 2016, pp. 393{398.
[9] E. Kochkina, M. Liakata, and I. Augenstein, \\Turing at semeval-2017 task 8:
Sequential approach to rumour stance classification with branch-lstm," arXiv
preprint arXiv:1704.07221, 2017.
[10] T. Chen, X. Li, H. Yin, and J. Zhang, \\Call attention to rumors: Deep attention based recurrent neural networks for early rumor detection," in PacificAsia Conference on Knowledge Discovery and Data Mining. Springer, 2018,
pp. 40{52.
[11] G. Giasemidis, C. Singleton, I. Agrafiotis, J. R. Nurse, A. Pilgrim, C. Willis,
and D. V. Greetham, \\Determining the veracity of rumours on twitter," in
International Conference on Social Informatics. Springer, 2016, pp. 185{205.
[12] C. Boididou, S. Papadopoulos, Y. Kompatsiaris, S. Schifferes, and N. Newman, \\Challenges of computational verification in social multimedia," in Proceedings of the 23rd International Conference on World Wide Web. ACM,
2014, pp. 743{748.
[13] J. Ma, W. Gao, P. Mitra, S. Kwon, B. J. Jansen, K.-F. Wong, and M. Cha,
\\Detecting rumors from microblogs with recurrent neural networks." in Ijcai,
2016, pp. 3818{3824.
[14] S. Kwon, M. Cha, and K. Jung, \\Rumor detection over varying time windows," PloS one, vol. 12, no. 1, p. e0168344, 2017.
[15] J. Chung, C. Gulcehre, K. Cho, and Y. Bengio, \\Empirical evaluation
of gated recurrent neural networks on sequence modeling," arXiv preprint
arXiv:1412.3555, 2014.
[16] C. Wardle and H. Derakhshan, \\Information disorder: Toward an interdisciplinary framework for research and policy making," Council of Europe Report,
vol. 27, 2017.
[17] A. Zubiaga, M. Liakata, R. Procter, G. W. S. Hoi, and P. Tolmie, \\Analysing
how people orient to and spread rumours in social media by looking at conversational threads," PloS one, vol. 11, no. 3, p. e0150989, 2016.
[18] A. Zubiaga, M. Liakata, and R. Procter, \\Exploiting context for rumour
detection in social media," in International Conference on Social Informatics.
Springer, 2017, pp. 109{123.
[19] R. Collobert and J. Weston, \\A unified architecture for natural language
processing: Deep neural networks with multitask learning," in Proceedings
of the 25th international conference on Machine learning. ACM, 2008, pp.
160{167.
[20] O. Enayet and S. R. El-Beltagy, \\Niletmrg at semeval-2017 task 8: Determining rumour and veracity support for rumours on twitter." in Proceedings
of the 11th International Workshop on Semantic Evaluation (SemEval-2017),
2017, pp. 470{474.
[21] R. Caruana, \\Multitask learning: A knowledge-based source of inductive bias.
machine learning," 1997.
[22] J. Baxter, \\A bayesian/information theoretic model of learning to learn via
multiple task sampling," Machine learning, vol. 28, no. 1, pp. 7{39, 1997.
[23] L. Duong, T. Cohn, S. Bird, and P. Cook, \\Low resource dependency parsing:
Cross-lingual parameter sharing in a neural network parser," in Proceedings
of the 53rd Annual Meeting of the Association for Computational Linguistics
and the 7th International Joint Conference on Natural Language Processing
(Volume 2: Short Papers), 2015, pp. 845{850.
[24] Y. Yang and T. M. Hospedales, \\Trace norm regularised deep multi-task
learning," arXiv preprint arXiv:1606.04038, 2016.
[25] A. Zubiaga, M. Liakata, and R. Procter, \\Learning reporting dynamics during breaking news for rumour detection in social media," arXiv preprint
arXiv:1610.07363, 2016.
[26] R. McCreadie, C. Macdonald, and I. Ounis, \\Crowdsourced rumour identification during emergencies," in Proceedings of the 24th International Conference
on World Wide Web. ACM, 2015, pp. 965{970.
[27] S. Ruder, \\An overview of multi-task learning in deep neural networks," arXiv
preprint arXiv:1706.05098, 2017.
[28] Y. S. Abu-Mostafa, \\Learning from hints in neural networks," Journal of
complexity, vol. 6, no. 2, pp. 192{198, 1990.
[29] J. Baxter, \\A model of inductive bias learning," Journal of artificial intelligence research, vol. 12, pp. 149{198, 2000.
[30] K. Cho, B. Van Merri¨enboer, D. Bahdanau, and Y. Bengio, \\On the properties
of neural machine translation: Encoder-decoder approaches," arXiv preprint
arXiv:1409.1259, 2014.
[31] D. Bahdanau, K. Cho, and Y. Bengio, \\Neural machine translation by jointly
learning to align and translate," arXiv preprint arXiv:1409.0473, 2014.
[32] T. Mikolov, K. Chen, G. Corrado, and J. Dean, \\Efficient estimation of word
representations in vector space," arXiv preprint arXiv:1301.3781, 2013.
描述 碩士
國立政治大學
資訊科學系碩士在職專班
105971005
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0105971005
資料類型 thesis
dc.contributor.advisor 胡毓忠zh_TW
dc.contributor.advisor Hu, Yuh-Jongen_US
dc.contributor.author (作者) 劉永鈞zh_TW
dc.contributor.author (作者) Liou, Yung-Jiunen_US
dc.creator (作者) 劉永鈞zh_TW
dc.creator (作者) Liou, Yung-Jiunen_US
dc.date (日期) 2020en_US
dc.date.accessioned 2-三月-2020 11:38:01 (UTC+8)-
dc.date.available 2-三月-2020 11:38:01 (UTC+8)-
dc.date.issued (上傳時間) 2-三月-2020 11:38:01 (UTC+8)-
dc.identifier (其他 識別碼) G0105971005en_US
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/128991-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 資訊科學系碩士在職專班zh_TW
dc.description (描述) 105971005zh_TW
dc.description.abstract (摘要) 偵測假新聞是一項十分艱鉅的任務，包含偵測假新聞(Rumour Detection)、假新聞追蹤(Rumour Tracking)及立場分類(Stance Classification)，從這些方法最終對假新聞作驗證(Rumour Verification)。欲做到辨識新聞的驗證以使讀者能閱讀到正確的新聞及資訊，本研究希望探索以多重任務學習(Multi-Task Learning, MTL)用於處理數量龐大的假新聞資料上，並比較與傳統深度學習的差異，達到自動辨識及判別假新聞的目的。
本研究使用RumourEval、PHEME兩種假新聞資料集來進行深度遞歸神經網路(Recurrent Neural Network, RNN)中的GRU（Gate Recurrent Unit）演算法實作，並進行多重任務學習的訓練，對假新聞進行分類，進而找出處理識別假新聞的最佳參數。最後透過各種模擬實驗來比較改良過後的深度學習演算法(即GRU)與傳統深度學習的差異，並依據實驗結果進行量化與質化的分析。		zh_TW
dc.description.abstract (摘要) Detecting fake news is a very difficult task, including Rumour Detection,Rumour Tracking and Stance Classification, and finally leading to Rumour Verification. To identify the authenticity of news so that readers can read the correct news and information, this research hopes to explore the use of Multi-Task Learning technology for processing a large number of fake news datasets and compare it with traditional deep learning, to achieve the purpose of automatically identifying and distinguishing fake news.
This research uses two fake news datasets, RumourEval and PHEME,to implement the GRU (Gated Recurrent Unit) algorithm of the Recurrent Neural Network (RNN), and trains for multiple tasks to perform fake news classification to find the best parameters for handling fake news. Finally, through various simulation experiments, the differences between the improved and traditional deep learning algorithm will be compared, and quantitative and qualitative analysis is performed based on the experimental results		en_US
dc.description.tableofcontents 目錄 iv
表目錄 vi
圖目錄 vii
第一章導論 1
1.1 研究動機 1
1.2 研究目的 2
第二章研究背景 4
2.1 假新聞定義 4
2.2 假新聞資料集 5
2.2.1 RumourEval 5
2.2.2 PHEME 6
2.3 多重任務學習 8
2.3.1 硬參數共享 9
2.3.2 軟參數共享 10
2.4 遞歸神經網路及GRU 10
2.4.1 遞歸神經網路 10
2.4.2 長短期記憶模型 12
2.4.3 GRU 13
第三章相關研究 14
3.1 假新聞資料集研究案例 14
3.2 假新聞偵測研究案例 16
3.3 多重任務學習研究案例 16
3.4 遞歸神經網路及GRU研究案例 18
第四章學習流程設計 19
4.1 假新聞分類模型 19
4.2 序列法 20
4.3 多重任務學習方法 22
4.4 特徵 22
第五章研究實作與比較 23
5.1 環境前處理流程 23
5.1.1 Keras 23
5.1.2 CUDA 24
5.2 實作使用技術及分類評價指標 24
5.3 實驗結果與比較 25
第六章結論與未來展望 28
6.1 研究結論 28
6.2 未來展望 28
參考文獻 29		zh_TW
dc.format.extent 1543537 bytes-
dc.format.mimetype application/pdf-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0105971005en_US
dc.subject (關鍵詞) 社交媒體zh_TW
dc.subject (關鍵詞) 假新聞zh_TW
dc.subject (關鍵詞) 錯誤資訊zh_TW
dc.subject (關鍵詞) 多重任務學習zh_TW
dc.subject (關鍵詞) 假新聞資料集zh_TW
dc.subject (關鍵詞) 遞歸神經網路zh_TW
dc.subject (關鍵詞) 傳統深度學習zh_TW
dc.subject (關鍵詞) Social Mediaen_US
dc.subject (關鍵詞) Fake Newsen_US
dc.subject (關鍵詞) Misinformationen_US
dc.subject (關鍵詞) Muti-Task Learningen_US
dc.subject (關鍵詞) PHEMEen_US
dc.subject (關鍵詞) Fake News Dataseten_US
dc.subject (關鍵詞) Recurrent Neural Networken_US
dc.subject (關鍵詞) GRUen_US
dc.subject (關鍵詞) Traditional Deep Learningen_US
dc.title (題名) 以深度遞歸神經網路實施多重任務學習偵測假新聞zh_TW
dc.title (題名) Deep Recurrent Neural Networks with Multi-Task Learning for Fake News Detectionen_US
dc.type (資料類型) thesisen_US
dc.relation.reference (參考文獻) [1] P. N. Howard, G. Bolsover, B. Kollanyi, S. Bradshaw, and L.-M. Neudert,
\\Junk news and bots during the us election: What were michigan voters
sharing over twitter," CompProp, OII, Data Memo, 2017.
[2] E. Kochkina, M. Liakata, and A. Zubiaga, \\All-in-one: Multi-task learning
for rumour verification," arXiv preprint arXiv:1806.03713, 2018.
[3] A. Zubiaga, A. Aker, K. Bontcheva, M. Liakata, and R. Procter, \\Detection and resolution of rumours in social media: A survey," ACM Computing
Surveys (CSUR), vol. 51, no. 2, p. 32, 2018.
[4] Z. Zhao, P. Resnick, and Q. Mei, \\Enquiring minds: Early detection of rumors
in social media from enquiry posts," in Proceedings of the 24th International
Conference on World Wide Web. International World Wide Web Conferences
Steering Committee, 2015, pp. 1395{1405.
[5] L. Derczynski, K. Bontcheva, M. Liakata, R. Procter, G. W. S. Hoi, and
A. Zubiaga, \\Semeval-2017 task 8: Rumoureval: Determining rumour veracity and support for rumours," arXiv preprint arXiv:1704.05972, 2017.
[6] M. Mendoza, B. Poblete, and C. Castillo, \\Twitter under crisis: Can we trust
what we rt?" in Proceedings of the first workshop on social media analytics.
ACM, 2010, pp. 71{79.
[7] R. Procter, F. Vis, and A. Voss, \\Reading the riots on twitter: methodological innovation for the analysis of big data," International journal of social
research methodology, vol. 16, no. 3, pp. 197{214, 2013.
[8] M. Lukasik, P. Srijith, D. Vu, K. Bontcheva, A. Zubiaga, and T. Cohn,
\\Hawkes processes for continuous time sequence classification: an application to rumour stance classification in twitter," in Proceedings of the 54th
Annual Meeting of the Association for Computational Linguistics (Volume 2:
Short Papers), vol. 2, 2016, pp. 393{398.
[9] E. Kochkina, M. Liakata, and I. Augenstein, \\Turing at semeval-2017 task 8:
Sequential approach to rumour stance classification with branch-lstm," arXiv
preprint arXiv:1704.07221, 2017.
[10] T. Chen, X. Li, H. Yin, and J. Zhang, \\Call attention to rumors: Deep attention based recurrent neural networks for early rumor detection," in PacificAsia Conference on Knowledge Discovery and Data Mining. Springer, 2018,
pp. 40{52.
[11] G. Giasemidis, C. Singleton, I. Agrafiotis, J. R. Nurse, A. Pilgrim, C. Willis,
and D. V. Greetham, \\Determining the veracity of rumours on twitter," in
International Conference on Social Informatics. Springer, 2016, pp. 185{205.
[12] C. Boididou, S. Papadopoulos, Y. Kompatsiaris, S. Schifferes, and N. Newman, \\Challenges of computational verification in social multimedia," in Proceedings of the 23rd International Conference on World Wide Web. ACM,
2014, pp. 743{748.
[13] J. Ma, W. Gao, P. Mitra, S. Kwon, B. J. Jansen, K.-F. Wong, and M. Cha,
\\Detecting rumors from microblogs with recurrent neural networks." in Ijcai,
2016, pp. 3818{3824.
[14] S. Kwon, M. Cha, and K. Jung, \\Rumor detection over varying time windows," PloS one, vol. 12, no. 1, p. e0168344, 2017.
[15] J. Chung, C. Gulcehre, K. Cho, and Y. Bengio, \\Empirical evaluation
of gated recurrent neural networks on sequence modeling," arXiv preprint
arXiv:1412.3555, 2014.
[16] C. Wardle and H. Derakhshan, \\Information disorder: Toward an interdisciplinary framework for research and policy making," Council of Europe Report,
vol. 27, 2017.
[17] A. Zubiaga, M. Liakata, R. Procter, G. W. S. Hoi, and P. Tolmie, \\Analysing
how people orient to and spread rumours in social media by looking at conversational threads," PloS one, vol. 11, no. 3, p. e0150989, 2016.
[18] A. Zubiaga, M. Liakata, and R. Procter, \\Exploiting context for rumour
detection in social media," in International Conference on Social Informatics.
Springer, 2017, pp. 109{123.
[19] R. Collobert and J. Weston, \\A unified architecture for natural language
processing: Deep neural networks with multitask learning," in Proceedings
of the 25th international conference on Machine learning. ACM, 2008, pp.
160{167.
[20] O. Enayet and S. R. El-Beltagy, \\Niletmrg at semeval-2017 task 8: Determining rumour and veracity support for rumours on twitter." in Proceedings
of the 11th International Workshop on Semantic Evaluation (SemEval-2017),
2017, pp. 470{474.
[21] R. Caruana, \\Multitask learning: A knowledge-based source of inductive bias.
machine learning," 1997.
[22] J. Baxter, \\A bayesian/information theoretic model of learning to learn via
multiple task sampling," Machine learning, vol. 28, no. 1, pp. 7{39, 1997.
[23] L. Duong, T. Cohn, S. Bird, and P. Cook, \\Low resource dependency parsing:
Cross-lingual parameter sharing in a neural network parser," in Proceedings
of the 53rd Annual Meeting of the Association for Computational Linguistics
and the 7th International Joint Conference on Natural Language Processing
(Volume 2: Short Papers), 2015, pp. 845{850.
[24] Y. Yang and T. M. Hospedales, \\Trace norm regularised deep multi-task
learning," arXiv preprint arXiv:1606.04038, 2016.
[25] A. Zubiaga, M. Liakata, and R. Procter, \\Learning reporting dynamics during breaking news for rumour detection in social media," arXiv preprint
arXiv:1610.07363, 2016.
[26] R. McCreadie, C. Macdonald, and I. Ounis, \\Crowdsourced rumour identification during emergencies," in Proceedings of the 24th International Conference
on World Wide Web. ACM, 2015, pp. 965{970.
[27] S. Ruder, \\An overview of multi-task learning in deep neural networks," arXiv
preprint arXiv:1706.05098, 2017.
[28] Y. S. Abu-Mostafa, \\Learning from hints in neural networks," Journal of
complexity, vol. 6, no. 2, pp. 192{198, 1990.
[29] J. Baxter, \\A model of inductive bias learning," Journal of artificial intelligence research, vol. 12, pp. 149{198, 2000.
[30] K. Cho, B. Van Merri¨enboer, D. Bahdanau, and Y. Bengio, \\On the properties
of neural machine translation: Encoder-decoder approaches," arXiv preprint
arXiv:1409.1259, 2014.
[31] D. Bahdanau, K. Cho, and Y. Bengio, \\Neural machine translation by jointly
learning to align and translate," arXiv preprint arXiv:1409.0473, 2014.
[32] T. Mikolov, K. Chen, G. Corrado, and J. Dean, \\Efficient estimation of word
representations in vector space," arXiv preprint arXiv:1301.3781, 2013.		zh_TW
dc.identifier.doi (DOI) 10.6814/NCCU202000256en_US