{"type":"ScholarlyArticle","title":"Gated POS-Level Language Model for Authorship Verification","authors":[{"name":"Linshu Ouyang","slug":"linshu-ouyang","given":"Linshu","family":"Ouyang","affiliation":"School of Cyber Security , University of Chinese Academy of Sciences , Beijing , China"},{"name":"Yongzheng Zhang","slug":"yongzheng-zhang","given":"Yongzheng","family":"Zhang","affiliation":"School of Cyber Security , University of Chinese Academy of Sciences , Beijing , China"},{"name":"Hui Liu","slug":"hui-liu","given":"Hui","family":"Liu","affiliation":"Institute of Information Engineering , Chinese Academy of Sciences , Beijing , China"},{"name":"Yige Chen","slug":"yige-chen","given":"Yige","family":"Chen","affiliation":"Institute of Information Engineering , Chinese Academy of Sciences , Beijing , China"},{"name":"Yipeng Wang","slug":"yipeng-wang","given":"Yipeng","family":"Wang","affiliation":"Institute of Information Engineering , Chinese Academy of Sciences , Beijing , China"}],"abstract":"Authorship verification is an important problem that has many applications. The state-of-the-art deep authorship verification methods typically leverage character-level language models to encode author-specific writing styles. However, they often fail to capture syntactic level patterns, leading to sub-optimal accuracy in cross-topic scenarios. Also, due to imperfect cross-author parameter sharing, it's difficult for them to distinguish author-specific writing style from common patterns, leading to data-inefficient learning. \r\n\r\nThis paper introduces a novel POS-level (Part of Speech) gated RNN based language model to effectively learn the author-specific syntactic styles. The author-agnostic syntactic information obtained from the POS tagger pre-trained on large external datasets greatly reduces the number of effective parameters of our model, enabling the model to learn accurate author-specific syntactic styles with limited training data. We also utilize a gated architecture to learn the common syntactic writing styles with a small set of shared parameters and let the author-specific parameters focus on each author's special syntactic styles. Extensive experimental results show that our method achieves significantly better accuracy than state-of-the-art competing methods, especially in cross-topic scenarios (over 5\\% in terms of AUC-ROC).","doi":"10.24963/ijcai.2020/557","url":"https://ijcai.org/proceedings/2020/557","pdf":"https://ijcai.org/proceedings/2020/0557.pdf","markdown":"https://ijcai.org/proceedings/2020/557.md","bibtex":"https://ijcai.org/proceedings/2020/bibtex/557","volume":{"slug":"2020","title":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2020","url":"https://ijcai.org/proceedings/2020/","year":2020,"location":"Yokohama","dates":"January 2021 (postponed from 11-17 July 2020)","editors":"Christian Bessiere","isbn":"978-0-9992411-6-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2020"},"section":"Natural Language Processing","track":"Main track","number":557,"pages":{"first":4025,"last":4031},"keywords":["Data Mining: Mining Text, Web, Social Media","Natural Language Processing: Other","Natural Language Processing: NLP Applications and Tools","Natural Language Processing: Natural Language Processing"],"videos":[{"label":"Video","url":"https://videos.ijcai20.org/24630/file_24630.m3u8","length_sec":233,"provider":"hls"}],"references":[{"position":1,"raw":"[Abbasi and Chen, 2008] Ahmed Abbasi and Hsinchun Chen. Writeprints: A stylometric approach to identity- level identification and similarity detection in cyberspace. ACM Transactions on Information Systems (TOIS), 26(2):7, 2008.","title":"Writeprints: A stylometric approach to identitylevel identification and similarity detection in cyberspace","authors":"Ahmed Abbasi, Hsinchun Chen","year":"2008","venue":"ACM Transactions on Information Systems (TOIS)","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Bagnall, 2015] Douglas Bagnall. Author identification us- ing multi-headed recurrent neural networks. arXiv preprint arXiv:1506.04891, 2015.","title":"Author identification using multi-headed recurrent neural networks","authors":"Douglas Bagnall","year":"2015","venue":null,"doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Bengio et al., 2003] Yoshua Bengio, Réjean Ducharme, Pascal Vincent, and Christian Jauvin. A neural probabilis- tic language model. Journal of machine learning research, 3(Feb):1137-1155, 2003.","title":"A neural probabilistic language model","authors":"Yoshua Bengio, Réjean Ducharme, Pascal Vincent, Christian Jauvin","year":"2003","venue":"Journal of machine learning research","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Bevendorff et al., 2019a] Janek Bevendorff, Matthias Ha- gen, Benno Stein, and Martin Potthast. Bias analysis and mitigation in the evaluation of authorship verification. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 6301-6306, 2019.","title":"Bias analysis and mitigation in the evaluation of authorship verification","authors":"Bevendorff","year":"2019","venue":"Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Bevendorff et al., 2019b] Janek Bevendorff, Martin Pot- thast, Matthias Hagen, and Benno Stein. Heuristic au- thorship obfuscation. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 1098-1108, 2019.","title":"Heuristic authorship obfuscation","authors":"Bevendorff","year":"2019","venue":"Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Graves, 2013] Alex Graves. Generating sequences with re- current neural networks. arXiv preprint arXiv:1308.0850, 2013.","title":"Generating sequences with recurrent neural networks","authors":"Alex Graves","year":"2013","venue":null,"doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Halvani et al., 2017] Oren Halvani, Christian Winter, and Lukas Graner. Authorship verification based on compression-models. arXiv preprint arXiv:1706.00516, 2017.","title":"Authorship verification based on compression-models","authors":"Oren Halvani, Christian Winter, Lukas Graner","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Halvani et al., 2018] Oren Halvani, Lukas Graner, and Inna Vogel. Authorship verification in the absence of explicit features and thresholds. In European Conference on Infor- mation Retrieval, pages 454-465. Springer, 2018.","title":"Authorship verification in the absence of explicit features and thresholds","authors":"Oren Halvani, Lukas Graner, Inna Vogel","year":"2018","venue":"European Conference on Information Retrieval","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Hitschler et al., 2017] Julian Hitschler, Esther van den Berg, and Ines Rehbein. Authorship attribution with con- volutional neural networks and pos-eliding. In Proceed- ings of the Workshop on Stylistic Variation, pages 53-58, 2017.","title":"Authorship attribution with convolutional neural networks and pos-eliding","authors":"Julian Hitschler, Esther Van Den, Ines Berg, Rehbein","year":"2017","venue":"Proceedings of the Workshop on Stylistic Variation","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Hürlimann et al., 2015] Manuela Hürlimann, Benno Weck, Esther van den Berg, Simon Suster, and Malvina Nis- sim. Glad: Groningen lightweight authorship detection. In CLEF (Working Notes), 2015.","title":"Glad: Groningen lightweight authorship detection","authors":"Manuela Hürlimann, Benno Weck, Esther Van Den, Simon Berg, Malvina Suster, Nissim","year":"2015","venue":"CLEF (Working Notes)","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Khmelev and Teahan, 2003] Dmitry V Khmelev and William J Teahan. A repetition based measure for verifi- cation of text collections and for text categorization. In Proceedings of the 26th annual international ACM SIGIR conference on Research and development in informaion retrieval, pages 104-110. ACM, 2003.","title":"A repetition based measure for verification of text collections and for text categorization","authors":"V Dmitry, William J Khmelev, Teahan","year":"2003","venue":"Proceedings of the 26th annual international ACM SIGIR conference on Research and development in informaion retrieval","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Koppel and Schler, 2004] Moshe Koppel and Jonathan Schler. Authorship verification as a one-class classi- fication problem. In Proceedings of the twenty-first international conference on Machine learning, page 62. ACM, 2004.","title":"Authorship verification as a one-class classification problem","authors":"Schler Koppel, Moshe Koppel, Jonathan Schler","year":"2004","venue":"Proceedings of the twenty-first international conference on Machine learning","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Manning et al., 2008] Christopher D Manning, Prabhakar Raghavan, and Hinrich Schütze. Introduction to informa- tion retrieval. Cambridge university press, 2008.","title":"Introduction to information retrieval","authors":"Prabhakar Christopher D Manning, Hinrich Raghavan, Schütze","year":"2008","venue":null,"doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Matthew and Ines, 2015] Honnibal Matthew and Montani Ines. spacy: Industrial-strength nlp. https://spacy.io/, 2015. [Version=2.2.3].","title":"Honnibal Matthew and Montani Ines. spacy: Industrial-strength nlp","authors":"Ines Matthew","year":"2015","venue":null,"doi":null,"url":"https://spacy.io/,2015.[Version=2.2.3","internal_url":null},{"position":15,"raw":"[Paszke et al., 2019] Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. Pytorch: An imperative style, high- performance deep learning library. In Advances in Neural Information Processing Systems, pages 8024-8035, 2019.","title":"Pytorch: An imperative style, highperformance deep learning library","authors":"Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga","year":"2019","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Qi et al., 2018] Peng Qi, Timothy Dozat, Yuhao Zhang, and Christopher D. Manning. Universal dependency parsing from scratch. In Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, pages 160-170, Brussels, Belgium, Octo- ber 2018. Association for Computational Linguistics.","title":"Universal dependency parsing from scratch","authors":"Peng Qi, Timothy Dozat, Yuhao Zhang, Christopher D Manning","year":"2018","venue":"Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Ruder et al., 2016] Sebastian Ruder, Parsa Ghaffari, and John G Breslin. Character-level and multi-channel con- volutional neural networks for large-scale authorship attri- bution. arXiv preprint arXiv:1609.06686, 2016.","title":"Character-level and multi-channel convolutional neural networks for large-scale authorship attribution","authors":"Sebastian Ruder, Parsa Ghaffari, John G Breslin","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Shrestha et al., 2017] Prasha Shrestha, Sebastian Sierra, Fabio A González, Manuel Montes, Paolo Rosso, and Thamar Solorio. Convolutional neural networks for au- thorship attribution of short texts. In Proceedings of the 15th Conference of the European Chapter of the Associa- tion for Computational Linguistics: Volume 2, Short Pa- pers, pages 669-674, 2017.","title":"Convolutional neural networks for authorship attribution of short texts","authors":"Shrestha","year":"2017","venue":"Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Stamatatos et al., 2015] Efstathios Stamatatos, Walter Daelemans, Ben Verhoeven, Patrick Juola, Aurelio Lopez- Lopez, Martin Potthast, and Benno Stein. Overview of the author identification task at pan 2015. In CLEF 2015 Working Notes Papers, 2015.","title":"Overview of the author identification task at pan 2015","authors":"Efstathios Stamatatos, Walter Daelemans, Ben Verhoeven, Patrick Juola, Aurelio Lopez-Lopez, Martin Potthast, Benno Stein","year":"2015","venue":"CLEF 2015 Working Notes Papers","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Stamatatos, 2009] Efstathios Stamatatos. A survey of mod- ern authorship attribution methods. Journal of the Amer- ican Society for information Science and Technology, 60(3):538-556, 2009.","title":"A survey of modern authorship attribution methods","authors":"Efstathios Stamatatos","year":"2009","venue":"Journal of the American Society for information Science and Technology","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Werbos and others, 1990] Paul J Werbos et al. Backpropa- gation through time: what it does and how to do it. Pro- ceedings of the IEEE, 78(10):1550-1560, 1990.","title":"Backpropagation through time: what it does and how to do it","authors":"Werbos Paul","year":"1990","venue":"Proceedings of the IEEE","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Zhao et al., 2006] Ying Zhao, Justin Zobel, and Phil Vines. Using relative entropy for authorship attribution. In Asia Information Retrieval Symposium, pages 92-105. Springer, 2006.","title":"Using relative entropy for authorship attribution","authors":"Ying Zhao, Justin Zobel, Phil Vines","year":"2006","venue":"Asia Information Retrieval Symposium","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}