A Hybrid AI Model for Fake News Detection: Leveraging FastText and LSTM for Kurdish and English
DOI:
https://doi.org/10.31530/cjnst.2025.1.1Keywords:
Fake news detection, Machine learning, Deep learning, LSTM, FastText, Kurdish language processing, TF-IDF, Hybrid modelAbstract
Background: The spread of misinformation on digital platforms has created a pressing need for effective fake news detection (FND), particularly in low-resource languages such as Kurdish.
Aims: This study aimed to develop and evaluate a hybrid FastText–LSTM model for fake news detection in Kurdish and English, and to compare its performance with traditional machine learning (ML) and deep learning (DL) approaches.
Methodology: The Kurdish Fake News Dataset (KDFND) was preprocessed, balanced, and used to train multiple models, including Logistic Regression, SVM, Random Forest, and LSTM. The proposed hybrid model combined FastText embeddings with LSTM architecture and was tested on both Kurdish and English text.
Results: The hybrid FastText–LSTM achieved superior accuracy, with 94.25% for Kurdish and 92.11% for English, outperforming traditional ML models and standalone LSTM. Balanced data significantly improved classification results.
Conclusion: The hybrid FastText–LSTM model provides an effective solution for fake news detection in both high- and low-resource languages, establishing a strong baseline for Kurdish. Future work should explore transformer-based architectures and multilingual detection strategies.
References
[1] D. A. Salh and R. M. Nabi, “Kurdish Fake News Detection Based on Machine Learning Approaches,” Passer J. Basic Appl. Sci., vol. 5, no. 2, pp. 262–271, 2023, doi: 10.24271/PSR.2023.380132.1226. DOI: https://doi.org/10.24271/psr.2023.380132.1226
[2] R. Azad, B. Mohammed, R. Mahmud, L. Zrar, and S. Sdiq, “Fake News Detection in low-resourced languages ‘Kurd-ish language’ using Machine learning algorithms,” Turkish J. Comput. Math. Educ. 4219 Res. Artic1., vol. 12, no. 6, pp. 4219–4225, 2021
[3] S. Bussa, A. Bodhankar, V. H. Patil, H. Pal, S. K. Bunkar, and A. R. K. Qureshi, “An Implementation of Machine Learning Algorithm for Fake News Detection,” Int. J. Re-cent Innov. Trends Comput. Commun., vol. 11, no. 3 (March), pp. 392–401, 2023, doi: 10.17762/ijritcc.v11i9s.7435. DOI: https://doi.org/10.17762/ijritcc.v11i9s.7435
[4] T. Saravanan, T. Jhaideep, and N. H. Bindu, “Detecting depression using Hybrid models created using Google’s BERT and Facebook’s Fast Text Algorithms,” 2022 2nd Int. Conf. Adv. Comput. Innov. Technol. Eng. ICACITE 2022, no. September, pp. 415–421, 2022, doi: 10.1109/ICACITE53722.2022.9823581. DOI: https://doi.org/10.1109/ICACITE53722.2022.9823581
[5] R. A. M. San Ahmed, “Hard Voting Approach using SVM, Naïve Bays and Decision Tree for Kurdish Fake News Detection,” Iraqi J. Comput. Sci. Math., vol. 4, no. 3, pp. 25–33, 2023, doi: 10.52866/ijcsm.2023.02.03.003. DOI: https://doi.org/10.52866/ijcsm.2023.02.03.003
[6] N. Sheikh, S. Parveen, A. Rehman, A. Naeem, M. Anjum, and M. Yasin, “Comparative Study of Fake News Detec-tion using Naïve Bayes and Logistic Model,” vol. 4883, no. July, pp. 1008–1015, 2024, doi: 10.53555/ks.v12i5.3390.
[7] S. Ahmadi, “KLPT – Kurdish Language Processing Toolkit,” pp. 72–84, 2020, doi: 10.18653/v1/2020.nlposs-1.11. DOI: https://doi.org/10.18653/v1/2020.nlposs-1.11
[8] Y. Taher, A. Moussaoui, and F. Moussaoui, “Automatic Fake News Detection based on Deep Learning, FastText and News Title,” Int. J. Adv. Comput. Sci. Appl., vol. 13, no. 1, pp. 146–158, 2022, doi: 10.14569/IJACSA.2022.0130118. DOI: https://doi.org/10.14569/IJACSA.2022.0130118
[9] L. Hu, S. Wei, Z. Zhao, and B. Wu, “Deep learning for fake news detection: A comprehensive survey,” AI Open, vol. 3, no. September, pp. 133–155, 2022, doi: 10.1016/j.aiopen.2022.09.001. DOI: https://doi.org/10.1016/j.aiopen.2022.09.001
[10] S. Sadiq, T. Aljrees, and S. Ullah, “Deepfake Detection on Social Media: Leveraging Deep Learning and FastText Embeddings for Identifying Machine-Generated Tweets,” IEEE Access, vol. 11, no. September, pp. 95008–95021, 2023, doi: 10.1109/ACCESS.2023.3308515. DOI: https://doi.org/10.1109/ACCESS.2023.3308515
[11] N. N. Prachi, M. Habibullah, M. E. H. Rafi, E. Alam, and R. Khan, “Detection of Fake News Using Machine Learning and Natural Language Processing Algorithms,” J. Adv. Inf. Technol., vol. 13, no. 6, pp. 652–661, 2022, doi: 10.12720/jait.13.6.652-661. DOI: https://doi.org/10.12720/jait.13.6.652-661
[12] T. S. Camelia and F. R. Fahim, “A Regularized LSTM Method for Detecting Fake News Articles,” IEEE, pp. 1–2, 2024. DOI: https://doi.org/10.1109/SPICSCON64195.2024.10941441
[13] E. Hashmi, S. Y. Yayilgan, M. M. Yamin, S. Ali, and M. Abomhara, “Advancing Fake News Detection: Hybrid Deep Learning With FastText and Explainable AI,” IEEE Access, vol. 12, no. March, pp. 44462–44480, 2024, doi: 10.1109/ACCESS.2024.3381038. DOI: https://doi.org/10.1109/ACCESS.2024.3381038
[14] H. Abdul, H. Al, H. Abdul, H. Al, and M. Jabardi, “De-tecting Fake News Using Machine Learning: A Compara-tive Study of Techniques,” J. Kufa Math. Comput., vol. 11, pp. 113–120, 2024. DOI: https://doi.org/10.31642/JoKMC/2018/110213
[15] Abubakr Salh, Dana; Nabi, Rebwar (2022), “Kurdish Dataset for Fake News Detection (KDFND)”, Mendeley Data, V1, doi: 10.17632/3zx9vpw3wh.1
[16] M. O. Arowolo, S. Misra, and R. O. Ogundokun, “A Machine Learning Technique for Detection of Social Media Fake News,” IJSWIS, vol. 19, no. 1, pp. 1–25, 2023, doi: 10.4018/IJSWIS.326120. DOI: https://doi.org/10.4018/IJSWIS.326120
[17] T. A. Rashid, A. M. Mustafa, and A. M. Saeed, “A Ro-bust Categorization System for Kurdish Sorani Text Doc-uments,” Inf. Technol. J., vol. 16, no. 1, pp. 27–34, 2016, doi: 10.3923/itj.2017.27.34. DOI: https://doi.org/10.3923/itj.2017.27.34
[18] I. Science, E. Don, S. Engineering, D. Bosco, and S. I. Science, “Deceptive Content Detection Using Machine Learning,” IJSREM, pp. 1–5, 2024, doi: 10.55041/IJSREM34830. DOI: https://doi.org/10.55041/IJSREM33974
[19] K. Yan, “Optimizing an English text reading recommen-dation model by integrating collaborative filtering algorithm and FastText classification method,” Heliyon, vol. 10, no. 9, p. e30413, 2024, doi: 10.1016/j.heliyon.2024.e30413. DOI: https://doi.org/10.1016/j.heliyon.2024.e30413
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Charmo Journal of Natural Sciences and Technologies
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
