A SHAP – Enhanced Voice-Based Conversational Agent for Agriculture Using BERT

Notice

This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.

Year : 2026 | Volume : 17 | 01 | Page :
    By

    Sirapani D,

  • Revathi J,

  • Ramadevi C,

  1. Student, Department of EEE, St. Joseph’s College of Engineering, Chennai, Tamil Nadu, India
  2. Student, Department of EEE, St. Joseph’s College of Engineering, Chennai, Tamil Nadu, India
  3. Assistant Professor, Department of EEE, St. Joseph’s College of Engineering, Chennai, Tamil Nadu, India

Abstract

The integration of advanced artificial intelligence technologies into modern agriculture has become increasingly important for narrowing the persistent knowledge gap faced by farmers, especially in regions with limited access to expert advisory services. While state-of-the-art language models such as BERT (Bidirectional Encoder Representations from Transformers) demonstrate exceptional performance in understanding and generating natural language, their opaque “black-box” nature often limits user confidence, trust, and widespread adoption. Farmers may hesitate to rely on recommendations when the reasoning behind system responses is unclear or difficult to interpret. To address this challenge, this paper presents a novel voice-enabled conversational agent specifically designed for the agricultural domain, built on a fine-tuned BERT model capable of accurately understanding and answering complex, context-rich farming queries. To overcome transparency and trust barriers, the proposed system integrates SHAP (SHapley Additive exPlanations), a widely recognized Explainable Artificial Intelligence (XAI) technique. SHAP provides clear, post-hoc explanations for each prediction by highlighting the key words and phrases that influenced the model’s response. This enables farmers to verify and understand the logic behind the advice provided, thereby fostering confidence in the system. Additionally, the conversational agent incorporates a voice-based interface that supports local languages, making it accessible to users with varying literacy levels. By combining robust natural language understanding, explainable reasoning, and user-friendly voice interaction, the system functions as a dependable digital advisory assistant. Ultimately, this approach has the potential to enhance decision-making, boost agricultural productivity, and support sustainable farming practices.

Keywords: Conversational Agent, BERT, Explainable AI (XAI), SHAP, Agriculture, Voice Interface, Natural Language Processing (NLP)

How to cite this article:
Sirapani D, Revathi J, Ramadevi C. A SHAP – Enhanced Voice-Based Conversational Agent for Agriculture Using BERT. Journal of Computer Technology & Applications. 2026; 17(01):-.
How to cite this URL:
Sirapani D, Revathi J, Ramadevi C. A SHAP – Enhanced Voice-Based Conversational Agent for Agriculture Using BERT. Journal of Computer Technology & Applications. 2026; 17(01):-. Available from: https://journals.stmjournals.com/jocta/article=2026/view=237233


References

  1. J Devlin J, Chang MW, Lee K, Toutanova K. Bert: Pre-training of deep bidirectional transformers for language understanding. InProceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies, volume 1 (long and short papers) 2019 Jun (pp. 4171-4186).
  2. Sarkar S, Babar MF, Hassan MM, Hasan M, Karmaker Santu SK. Processing Natural Language on Embedded Devices: How Well Do Transformer Models Perform?. InProceedings of the 15th ACM/SPEC International Conference on Performance Engineering 2024 May 7 (pp. 211-222).
  3. Chen H, Jiao F, Li X, Qin C, Ravaut M, Zhao R, Xiong C, Joty S. Chatgpt’s one-year anniversary: are open- source large language models catching up?. arXiv preprint arXiv:2311.16989. 2023 Nov 28.
  4. Wang D, Liu C, Zhao Z, Shen S, Liu L, Li B, Hu H, Wu M, Lin L, Zhao X, Wang X. Gujibert and gujigpt: Construction of intelligent information processing foundation language models for ancient texts. arXiv preprint arXiv:2307.05354. 2023 Jul 11.
  5. Chen T, Lv L, Wang D, Zhang J, Yang Y, Zhao Z, Wang C, Guo X, Chen H, Wang Q, Xu Y. Revolutionizing agrifood systems with artificial intelligence: a survey. arXiv preprint arXiv:2305.01899. 2023.
  6. Usip PU, Udo EN, Asuquo DE, James OR. A machine learning-based mobile chatbot for crop farmers. InInternational Conference on Electronic Governance with Emerging Technologies 2022 Sep 12 (pp. 192- 211). Cham: Springer Nature Switzerland.
  7. Gardezi M, Joshi B, Rizzo DM, Ryan M, Prutzer E, Brugler S, Dadkhah A. Artificial intelligence in farming: Challenges and opportunities for building trust. Agronomy Journal. 2024 May;116(3):1217-28.
  8. Min W, Zhou P, Xu L, Liu T, Li T, Huang M, Jin Y, Yi Y, Wen M, Jiang S, Jain R. From plate to production: Artificial intelligence in modern consumer-driven food systems. arXiv preprint arXiv:2311.02400. 2023 Nov 4.
  9. Salih AM, Raisi‐Estabragh Z, Galazzo IB, Radeva P, Petersen SE, Lekadir K, Menegaz G. A perspective on explainable artificial intelligence methods: SHAP and LIME. Advanced Intelligent Systems. 2025 Jan;7(1):2400304.
  10. Lundberg SM, Lee SI. A unified approach to interpreting model predictions. Advances in neural information processing systems. 2017;30.
  11. Zhao H, Chen H, Yang F, Liu N, Deng H, Cai H, Wang S, Yin D, Du M. Explainability for large language models: A survey. ACM Transactions on Intelligent Systems and Technology. 2024 Feb 22;15(2):1-38.
  12. Vainio-Pekka H, Agbese MO, Jantunen M, Vakkuri V, Mikkonen T, Rousi R, Abrahamsson P. The role of explainable AI in the research field of AI ethics. ACM Transactions on Interactive Intelligent Systems. 2023 Dec 8;13(4):1-39.
  13. Qasim M, Zia HB, Athar A, Habib T, Raza AA. Personalized weather information for low-literate farmers using multimodal dialog systems. International Journal of Speech Technology. 2021 Jun;24(2):455-71.
  14. Pratap V, Tjandra A, Shi B, Tomasello P, Babu A, Kundu S, Elkahky A, Ni Z, Vyas A, Fazel-Zarandi M, Baevski A. Scaling speech technology to 1,000+ languages. Journal of Machine Learning Research. 2024;25(97):1-52.
  15. Ngulube P. Leveraging information and communication technologies for sustainable agriculture and environmental protection among smallholder farmers in tropical Africa. Discover Environment. 2025 Jan 23;3(1):9.
Ahead of Print Subscription Original Research
Volume 17
01
Received 18/12/2025
Accepted 07/01/2026
Published 20/02/2026
Publication Time 64 Days
124

Login


My IP

PlumX Metrics