Tekrarlayan Sinir Ağı Tabanlı Derin Öğrenme Yaklaşımları ile Sahte Haber Tespiti ve Sınıflandırması

Year 2024, , 973 - 993, 25.06.2024

Halit Çetiner

https://doi.org/10.47495/okufbed.1199738

Abstract

Dünyada olup biten olaylar son kullanıcıya haber mecrası aracılığıyla aktarılmaktadır. Haberlerden aktarılan bilgilerin genellikle doğru olduğu düşünülmektedir. Ancak haber kanallarında dolaşan bilgilerde hata ya da yalan olabilmektedir. Aynı zamanda bu haberlerin ekonomi gibi ciddi ortamlarda etkisi de bulunmaktadır. Veri paylaşımının artış gösterdiği sosyal ağlarda haber verileri kontrolsüz bir şekilde yığılmaktadır. Bu veri yığınları içerisinde gerçek bilgiler olduğu gibi gerçek olmayan ticari, siyasi ya da satış hedefli farklı bilgilerde bulunmaktadır. Gerçek olmayan bilgiler, kullanıcılar tarafından paylaşılması sonucunda sahte bilgi ve veriler daha hızlı bir şekilde genişlemektedir. Bu tür haberler kullanıcıları doğrudan etkileyerek hatalı işlem yapmaya, yanlış bilgi sahibi olmaya veya maddi bir kayba neden olmaktadır. Belirtilen sebeplerden dolayı bu makalede doğal dil işleme Tekrarlayan Sinir Ağı (TSA) tabanlı derin öğrenme yöntemleri ile birleştirerek otomatik sahte haber sınıflandırma sistemleri önerilmiştir. Önerilen sistemler genel performans metrikleri kullanılarak 23,481 adet sahte haber, 21,417 adet gerçek haber içeren bir veri setinde test edilmiştir. Yapılan test işlemleri sonucunda önerilen BiLSTM yöntemi %99,72 doğruluk oranı sağlarken, önerilen GRU yöntemi %97,50 doğruluk oranına ulaşmıştır.

Keywords

Sahte Haber Sınıflandırma, BiLSTM, GRU, Convolutional Neural Network, Natural Language Processing

Fake News Detection and Classification with Recurrent Neural Network Based Deep Learning Approaches

Abstract

Events happening in the world are transmitted to the end user through the news channel. The information transmitted from the news is generally considered to be accurate. However, there may be errors or lies in the information that circulates on the news channels. At the same time, this news has an impact on serious environments, such as the economy. In social networks where data sharing is increasing, news data is piling up uncontrollably. In these data piles, there is real information as well as different information that is not real commercial, political, or sales-orientated. False information and data expand faster as a result of sharing false information by users. This news directly affects users, causing erroneous transactions, misinformation, or financial loss. For the stated reasons, automatic fake news classification systems are proposed in this article by combining natural language processing with Recurrent Neural Network (RNN) based deep learning methods. The proposed systems were tested on a dataset containing 23,481 fake news and 21,417 real news using general performance metrics. As a result of the test processes, the proposed BiLSTM method provided 99,72% accuracy, while the proposed GRU method accessed 97,50% accuracy.

Keywords

Convolutional Neural Network, Fake News Classification, BiLSTM, GRU, Natural Language Processing

References

• Aggarwal CC. Neural networks and deep learning. Springer 2018; 10: 973–978.
• Ahmed H, Traore I, Saad S. Detecting opinion spams and fake news using text classification. Security and Privacy 2018; 1(1): e9.
• Ahmed H, Traore I, Saad S. Detection of online fake news using n-gram analysis and machine learning techniques. International Conference on Intelligent, Secure, and Dependable Systems in Distributed And Cloud Environments 2017; 127–138.
• Almuzaini HA, Azmi AM. Impact of stemming and word embedding on deep learning-based arabic text categorization. IEEE Access 2020; 8: 127913–127928.
• Bakshy E, Messing S, Adamic LA. Exposure to ideologically diverse news and opinion on Facebook. American Association for the Advancement of Science 2015; 348(6239): 1130–1132.
• Bali APS, Fernandes M, Choubey S, Goel M. Comparative performance of machine learning algorithms for fake news detection BT - Advances in Computing and Data Sciences. In: Singh M, Gupta PK, Tyagi V, Flusser J, Ören T and Kashyap R (eds) 2019; 420–430.
• Balmas M. When fake news becomes real: Combined exposure to multiple news sources and political attitudes of inefficacy, alienation, and cynicism. Communication Research Sage Publications Sage CA: Los Angeles 2014; 41(3): 430–454.
• Barthel M, Mitchell A, Holcomb J. Many Americans believe fake news is sowing confusion. Pew Research Center 2016.
• Çetiner H, Kara B. Recurrent neural network based model development for wheat yield forecasting. Journal of Engineering Sciences of Adiyaman University 2022; 9(16): 204–218.
• Chen T, Xu R, He Y, Wang X. Improving sentiment analysis via sentence type classification using BiLSTM-CRF and CNN. Expert Systems with Applications 2017; 72: 221–230.
• Conroy NK, Rubin VL, Chen Y. Automatic deception detection: Methods for finding fake news. Proceedings of the Association for Information Science and Technology 2015; 52(1): 1–4.
• Feng S, Banerjee R, Choi Y. Syntactic stylometry for deception detection. Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics 2012; 171–175.
• Guacho GB, Abdali S, Shah N, Papalexakis EE. Semi-supervised content-based detection of misinformation via tensor embeddings. 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) 2018; 322–325.
• Gulli A, Pal S. Deep learning with Keras. Packt Publishing Ltd 2017.
• Hochreiter S, Schmidhuber J. Long short-term memory. Neural computation 1997; 9(8): 1735–1780.
• Horne B, Adali S. This just in: Fake news packs a lot in title, uses simpler, repetitive content in text body, more similar to satire than real news. Proceedings of the International AAAI Conference on Web and Social Media 2017; 759–766.
• Hosseinimotlagh S, Papalexakis EE. Unsupervised content-based identification of fake news articles with tensor decomposition ensembles. Proceedings of the Workshop on Misinformation and Misbehavior Mining on the Web (MIS2) 2018.
• Hu L, Wang C, Ye Z, Wang S. Estimating gaseous pollutants from bus emissions: A hybrid model based on GRU and XGBoost. Science of The Total Environment 2021; 783: 146870.
• Karimi H, Roy P, Saba-Sadiya S, Tang J. Multi-source multi-class fake news detection. Proceedings of the 27th International Conference on Computational Linguistics 2018; 1546–1557.
• Kaur S, Kumar P, Kumaraguru P. Automating fake news detection system using multi-level voting model. Soft Computing 2020; 24(12): 9049–9069.
• Kingma D, Ba J. Adam: A Method for Stochastic Optimization. International Conference on Learning Representations 2014.
• Kishwar A, Zafar A. Fake news detection on Pakistani news using machine learning and deep learning. Expert Systems with Applications 2023; 211: 118558.
• Li S. Application of recurrent neural networks in toxic comment classification. UCLA 2018.
• Liu G, Guo J. Bidirectional LSTM with attention mechanism and convolutional layer for text classification. Neurocomputing 2019; 337: 325–338.
• Nasir JA, Khan OS, Varlamis I. Fake news detection: A hybrid CNN-RNN based deep learning approach. International Journal of Information Management Data Insights 2021; 1(1): 100007.
• Ngada O, Haskins B. Fake news detection using content-based features and machine learning. 2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE) 2020; 1–6.
• Niu X, Hou Y, Wang P. Bi-directional LSTM with quantum attention mechanism for sentence modeling. International Conference on Neural Information Processing 2017; 178–188.
• Nowak J, Taspinar A, Scherer R. LSTM recurrent neural networks for short text and sentiment classification. International Conference on Artificial Intelligence and Soft Computing 2017; 553–562.
• Ozbay FA, Alatas B. Fake news detection within online social media using supervised artificial intelligence algorithms. Physica A: Statistical Mechanics and its Applications 2020; 540: 123174.
• Pang Z, Niu F, O’Neill Z. Solar radiation prediction using recurrent neural network and artificial neural network: A case study with comparisons. Renewable Energy 2020; 156: 279–289.
• Reis JCS, Correia A, Murai F, Veloso A, Benevenuto F. Supervised Learning for Fake News Detection. IEEE Intelligent Systems 2019; 34(2): 76–81.
• Sabeeh V, Zohdy M, Mollah A, Al Bashaireh R. Fake news detection on social media using deep learning and semantic knowledge sources. International Journal of Computer Science and Information Security (IJCSIS) 2020; 18(2).
• Sahoo SR, Gupta BB. Multiple features based approach for automatic fake news detection on social networks using deep learning. Applied Soft Computing 2021; 100: 106983.
• Sahoo SR, Gupta BB. Hybrid approach for detection of malicious profiles in twitter. Computers & Electrical Engineering 2019; 76: 65–81.
• Sunstein C. On Rumors How Falsehoods Spread, Why We Believe Them, What Can Be Done. Princeton University Press publisher, 2009.
• Tacchini E, Ballarin G, Della Vedova ML, Moret S, de Alfaro L. Some like it hoax: Automated fake news detection in social networks. arXiv preprint arXiv:1704.07506 2017.
• Wang J, Zhang Y, Yu L-C, Zhang X. Contextual sentiment embeddings via bi-directional GRU language model. Knowledge-Based Systems 2022; 235: 107663.
• Wang Y, Ma F, Jin Z, Yuan Y, Xun G, Jha K, Su L, Gao J. Eann: Event adversarial neural networks for multi-modal fake news detection. Proceedings of the 24th Acm Sigkdd International Conference on Knowledge Discovery & Data Mining 2018; 849–857.
• Wu L, Liu H. Tracing fake-news footprints: Characterizing social media messages by how they propagate. Proceedings of The Eleventh ACM International Conference on Web Search and Data Mining 2018; 637–645.
• Yang Y, Zheng L, Zhang J, Cui Q, Li Z, Yu PS. TI-CNN: Convolutional neural networks for fake news detection. arXiv preprint arXiv:1806.00749 2018.
• Zhang Y, Zhang Z, Miao D, Wang J. Three-way enhanced convolutional neural networks for sentence-level sentiment classification. Information Sciences 2019; 477: 55–64.