Stance Classification in Afan Oromo Using Deep Learning Approaches

Dejene Teka

doi:doi:10.11648/j.mlr.20251002.16

Research Article |

| Peer-Reviewed

Stance Classification in Afan Oromo Using Deep Learning Approaches

Dejene Teka^*

Published in Machine Learning Research (Volume 10, Issue 2)

Received: 6 August 2025 Accepted: 16 August 2025 Published: 19 December 2025

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

Stance detection is an important task in natural language processing (NLP) that seeks to determine a speaker’s or writer’s position toward a given topic. While substantial progress has been achieved for major languages, low-resource languages such as Afan Oromo remain largely underexplored. This study introduces a deep learning–based approach for stance detection in Afan Oromo, leveraging a newly collected and annotated dataset of over one million sentences from social media platforms, particularly Facebook. Three deep learning models—Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (Bi-LSTM)—were implemented and evaluated. Among these, CNN achieved the highest accuracy of 85.9%, outperforming LSTM (81.4%) and Bi-LSTM (79.8%). The superior performance of CNN is attributed to its ability to capture local spatial features in text, which is particularly beneficial for short, informal social media posts. These results demonstrate the feasibility and effectiveness of deep learning techniques for stance detection in low-resource languages. Furthermore, the findings contribute to advancing language technologies for Afan Oromo and open pathways for future research in social media analysis, sentiment monitoring, and political discourse understanding in local contexts.

Published in	Machine Learning Research (Volume 10, Issue 2)
DOI	10.11648/j.mlr.20251002.16
Page(s)	151-157
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2025. Published by Science Publishing Group

Keywords

Stance Detection, Afan Oromo, CNN, LSTM, Bi-LSTM, Deep Learning, Low-Resource Languages, Natural Language Processing

References

[1]	Wondimu. Tegegne, The development of written Afan Oromo and the appropriateness of Qubee, Latin Script, for Afan Oromo writing. Historical Research Letter, p. 28, 2016.
[2]	Abdo Ababor. Abafogi, "Enhanced Word Sense Disambiguation Algorithm for Afaan Oromoo.," Information Engineering and Electronic Business, pp. 41-50, February 8, 2023. https://doi.org/10.5815/ijieeb.2023.01.04
[3]	Saif M., Parinaz Sobhani, and Svetlana Kiritchenko. Mohammad, "Stance and sentiment in tweets," Transactions on Internet Technology (TOIT), pp. 1-23, 2017. https://doi.org/10.48550/arXiv.1605.01655
[4]	A. Tazeb, "The Framing of Political Uprising on Social Media: The Case of," Journalism and Communication, pp. 1-120, 2017.
[5]	D. Küçük and F. C. SIGIR, "Stance detection: Concepts, approaches, resources," Proceedings of the 44th International ACM, 2021.
[6]	"Stance detection D. Küçük and F. C. SIGIR, "Concepts, approaches, resources," Proceedings of the 44th International ACM;, 2021.
[7]	S. O. Edosomwan, "The history of social media and its impact on business," Article in The Journal of Applied Management & Entrepreneurship, vol. 16, 2011.
[8]	A. Arora, P. Nakov and I. Augenstein M. Hardalov, "ASurvey on Stance Detection for Mis- and Disinformation Identification," arXiv preprint, pp. 1-19, 8 May 2022.
[9]	D. Dou and D. Lowd J. Ebrahimi, "A joint sentiment-target-stance model for stance classification in tweets," Proceedings of COLING 2016, the 26th, 2016.
[10]	H. Luqman and M. Ahmed N. Alturayeif, "A systematic review of machine learning techniques for stance detection and its applications," Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2016.
[11]	T. Tuunanen, C. E. Gengler, M. Rossi, W. Hui, V. Virtanen and J. K. Peffers, "Design Science Research Process: A Model for Producing and," International Conference on Design Science Research in Information Systems, https://doi.org/10.2753/MIS0742-1222240302
[12]	Kareem, et al. Darwish, "Predicting online islamophobic behavior after# ParisAttacks.," The Journal of Web Science, vol. 3, pp. 34-52, 2018.
[13]	D. Küçük and F. Can, "Stance detection: A survey," Journal of ACM Computing, vol. 53, pp. 1-37, 2021.
[14]	Galataa. Abbabaa, "QAACCESSA QABIYYEE FAARUU QAALLUU MACCAA: AANAA SIBUU SIREETTI. Diss. Yuunivarsiitii Haramaayaa," 2018.
[15]	Tom, et al. Young, "Recent trends in deep learning based natural language processing," IEEE, pp. 55-77, 2018.
[16]	Nader, et al. Tavaf, "GRAPPA-GANs for parallel MRI reconstruction." arXiv preprint arXiv: 2101.03135 (2021).".
[17]	Maryem, et al Rhanoui, "A CNN-BiLSTM model for document-level sentiment analysis," Machine Learning and Knowledge Extraction, pp. 832-847., 2019.
[18]	Yann, Yoshua Bengio, and Geoffrey Hinton LeCun, "Deep learning," pp. 436-444, 2015.
[19]	S. M., & Bravo-Marquez, F. Mohammad, "Stance and sentiment in tweets. ACM Transactions on Internet Technology (TOIT), 17(3), 26.," 2017.
[20]	A., & Paul, S. Bhatia, "Stance classification in online debate portal," In 2016 IEEE International Conference on Computational Intelligence and Computing Research (ICCIC), pp. 1-5, 2016.
[21]	C., Huang, L., Qiu, X., & Hu, X. Sun, "LSTM-based deep learning models for non-factoid answer selection. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics," Long Papers, pp. 1142-1151.
[22]	Y., Du, N., Yuan, M., & Huang, X. Li, "A Survey on Bi-Directional LSTM: Architecture, Modeling, and Applications," Journal of Ocean University of China, pp. 513-524., 2019.
[23]	D., Wang, X., & Fang, Y. Zhang, "A comprehensive survey on convolutional neural networks." Journal of Machine Learning Research, pp. 3199-3230., 2017.
[24]	T., Chen, K., Corrado, G., & Dean, J. Mikolov, "Efficient estimation of word representations in vector space," 2013.

Cite This Article

Plain Text BibTeX RIS

APA Style

Teka, D. (2025). Stance Classification in Afan Oromo Using Deep Learning Approaches. Machine Learning Research, 10(2), 151-157. https://doi.org/10.11648/j.mlr.20251002.16

Copy | Download

ACS Style

Teka, D. Stance Classification in Afan Oromo Using Deep Learning Approaches. Mach. Learn. Res. 2025, 10(2), 151-157. doi: 10.11648/j.mlr.20251002.16

Copy | Download

AMA Style

Teka D. Stance Classification in Afan Oromo Using Deep Learning Approaches. Mach Learn Res. 2025;10(2):151-157. doi: 10.11648/j.mlr.20251002.16

Copy | Download

@article{10.11648/j.mlr.20251002.16,
  author = {Dejene Teka},
  title = {Stance Classification in Afan Oromo Using Deep Learning Approaches},
  journal = {Machine Learning Research},
  volume = {10},
  number = {2},
  pages = {151-157},
  doi = {10.11648/j.mlr.20251002.16},
  url = {https://doi.org/10.11648/j.mlr.20251002.16},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.mlr.20251002.16},
  abstract = {Stance detection is an important task in natural language processing (NLP) that seeks to determine a speaker’s or writer’s position toward a given topic. While substantial progress has been achieved for major languages, low-resource languages such as Afan Oromo remain largely underexplored. This study introduces a deep learning–based approach for stance detection in Afan Oromo, leveraging a newly collected and annotated dataset of over one million sentences from social media platforms, particularly Facebook. Three deep learning models—Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (Bi-LSTM)—were implemented and evaluated. Among these, CNN achieved the highest accuracy of 85.9%, outperforming LSTM (81.4%) and Bi-LSTM (79.8%). The superior performance of CNN is attributed to its ability to capture local spatial features in text, which is particularly beneficial for short, informal social media posts. These results demonstrate the feasibility and effectiveness of deep learning techniques for stance detection in low-resource languages. Furthermore, the findings contribute to advancing language technologies for Afan Oromo and open pathways for future research in social media analysis, sentiment monitoring, and political discourse understanding in local contexts.},
 year = {2025}
}

Copy | Download

TY  - JOUR
T1  - Stance Classification in Afan Oromo Using Deep Learning Approaches
AU  - Dejene Teka
Y1  - 2025/12/19
PY  - 2025
N1  - https://doi.org/10.11648/j.mlr.20251002.16
DO  - 10.11648/j.mlr.20251002.16
T2  - Machine Learning Research
JF  - Machine Learning Research
JO  - Machine Learning Research
SP  - 151
EP  - 157
PB  - Science Publishing Group
SN  - 2637-5680
UR  - https://doi.org/10.11648/j.mlr.20251002.16
AB  - Stance detection is an important task in natural language processing (NLP) that seeks to determine a speaker’s or writer’s position toward a given topic. While substantial progress has been achieved for major languages, low-resource languages such as Afan Oromo remain largely underexplored. This study introduces a deep learning–based approach for stance detection in Afan Oromo, leveraging a newly collected and annotated dataset of over one million sentences from social media platforms, particularly Facebook. Three deep learning models—Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (Bi-LSTM)—were implemented and evaluated. Among these, CNN achieved the highest accuracy of 85.9%, outperforming LSTM (81.4%) and Bi-LSTM (79.8%). The superior performance of CNN is attributed to its ability to capture local spatial features in text, which is particularly beneficial for short, informal social media posts. These results demonstrate the feasibility and effectiveness of deep learning techniques for stance detection in low-resource languages. Furthermore, the findings contribute to advancing language technologies for Afan Oromo and open pathways for future research in social media analysis, sentiment monitoring, and political discourse understanding in local contexts.
VL  - 10
IS  - 2
ER  -

Copy | Download

Author Information

Dejene Teka

Department of Information Technology, Mattu University, Oromia, Ethiopia

Contact Email

http://orcid.org/0009-0002-3674-3773

Download PDF

Submit an Article

Sections

Plain Text BibTeX RIS

APA Style

Teka, D. (2025). Stance Classification in Afan Oromo Using Deep Learning Approaches. Machine Learning Research, 10(2), 151-157. https://doi.org/10.11648/j.mlr.20251002.16

Copy | Download

ACS Style

Teka, D. Stance Classification in Afan Oromo Using Deep Learning Approaches. Mach. Learn. Res. 2025, 10(2), 151-157. doi: 10.11648/j.mlr.20251002.16

Copy | Download

AMA Style

Teka D. Stance Classification in Afan Oromo Using Deep Learning Approaches. Mach Learn Res. 2025;10(2):151-157. doi: 10.11648/j.mlr.20251002.16

Copy | Download

@article{10.11648/j.mlr.20251002.16,
  author = {Dejene Teka},
  title = {Stance Classification in Afan Oromo Using Deep Learning Approaches},
  journal = {Machine Learning Research},
  volume = {10},
  number = {2},
  pages = {151-157},
  doi = {10.11648/j.mlr.20251002.16},
  url = {https://doi.org/10.11648/j.mlr.20251002.16},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.mlr.20251002.16},
  abstract = {Stance detection is an important task in natural language processing (NLP) that seeks to determine a speaker’s or writer’s position toward a given topic. While substantial progress has been achieved for major languages, low-resource languages such as Afan Oromo remain largely underexplored. This study introduces a deep learning–based approach for stance detection in Afan Oromo, leveraging a newly collected and annotated dataset of over one million sentences from social media platforms, particularly Facebook. Three deep learning models—Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (Bi-LSTM)—were implemented and evaluated. Among these, CNN achieved the highest accuracy of 85.9%, outperforming LSTM (81.4%) and Bi-LSTM (79.8%). The superior performance of CNN is attributed to its ability to capture local spatial features in text, which is particularly beneficial for short, informal social media posts. These results demonstrate the feasibility and effectiveness of deep learning techniques for stance detection in low-resource languages. Furthermore, the findings contribute to advancing language technologies for Afan Oromo and open pathways for future research in social media analysis, sentiment monitoring, and political discourse understanding in local contexts.},
 year = {2025}
}

Copy | Download

TY  - JOUR
T1  - Stance Classification in Afan Oromo Using Deep Learning Approaches
AU  - Dejene Teka
Y1  - 2025/12/19
PY  - 2025
N1  - https://doi.org/10.11648/j.mlr.20251002.16
DO  - 10.11648/j.mlr.20251002.16
T2  - Machine Learning Research
JF  - Machine Learning Research
JO  - Machine Learning Research
SP  - 151
EP  - 157
PB  - Science Publishing Group
SN  - 2637-5680
UR  - https://doi.org/10.11648/j.mlr.20251002.16
AB  - Stance detection is an important task in natural language processing (NLP) that seeks to determine a speaker’s or writer’s position toward a given topic. While substantial progress has been achieved for major languages, low-resource languages such as Afan Oromo remain largely underexplored. This study introduces a deep learning–based approach for stance detection in Afan Oromo, leveraging a newly collected and annotated dataset of over one million sentences from social media platforms, particularly Facebook. Three deep learning models—Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (Bi-LSTM)—were implemented and evaluated. Among these, CNN achieved the highest accuracy of 85.9%, outperforming LSTM (81.4%) and Bi-LSTM (79.8%). The superior performance of CNN is attributed to its ability to capture local spatial features in text, which is particularly beneficial for short, informal social media posts. These results demonstrate the feasibility and effectiveness of deep learning techniques for stance detection in low-resource languages. Furthermore, the findings contribute to advancing language technologies for Afan Oromo and open pathways for future research in social media analysis, sentiment monitoring, and political discourse understanding in local contexts.
VL  - 10
IS  - 2
ER  -

Copy | Download