EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Deep Learning-Based Multilingual Smart Farming System for Crop Recommendation and Nutrient Monitoring

M. K. Shriwas, K. Sindhi, P. Thakre

AFRICAN JOURNAL OF APPLIED RESEARCH · 04.07.2026 · 12 (5) · 260-280

DOI: 10.26437/arzmee72 ↗

Kurz gefasst

Fachartikel in der Zeitschrift AFRICAN JOURNAL OF APPLIED RESEARCH, online erschienen am 04.07.2026. Verfasst von M. K. Shriwas und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 16 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

The objective of this study is to develop a deep learning-based multilingual smart farming system for crop recommendation and nutrient monitoring in Maharashtra. The study aims to support farmers in selecting suitable crops by considering soil type, weather conditions, nutrient parameters, and regional agricultural factors, while also improving accessibility through advisory support in Marathi, Hindi, and English. Design / Methodology / Approach: The proposed framework integrates soil image classification, weather and nutrient feature analysis, and crop recommendation into a unified decision-support system. A CNN-based model is used for soil type classification, while a deep neural network is employed for crop recommendation using environmental, soil, seasonal, district-level, and nutrient-related parameters. Research Limitation: The major limitation of the adopted methodology is that the system is evaluated using available soil images and tabular weather–nutrient datasets, rather than through large-scale, real-time field deployment. The framework does not yet include live IoT-based nutrient sensing, long-term seasonal field validation, or direct farmer feedback from practical use.

Findings

The experimental analysis shows that the proposed multilingual deep learning-based model outperforms conventional machine learning approaches in accuracy, precision, recall, F1-score, Top-3 recommendation accuracy, and computation time. The results indicate that combining soil imagery with weather and nutrient features improves the reliability of crop recommendations and supports more effective agricultural decision-making. Practical Implication: The proposed system can assist farmers, agricultural advisors, and decision-support agencies by providing crop recommendations and nutrient-related guidance in regional languages. This can improve nutrient management, reduce dependency on manual advisory practices, and support region-specific smart farming in Maharashtra. Social Implication: As a decision-support system embedded within farming communities, its adoption reshapes livelihoods, knowledge systems, labour dynamics, and rural development trajectories. This section critically examines these dimensions. Originality / Value: The originality of this work lies in integrating soil classification, weather–nutrient analysis, crop recommendation, and multilingual advisory generation into a single deep learning-based framework for agriculture in Maharashtra.

Zitieren

APA

Shriwas, M. K., Sindhi, K., & Thakre, P. (2026). Deep Learning-Based Multilingual Smart Farming System for Crop Recommendation and Nutrient Monitoring. AFRICAN JOURNAL OF APPLIED RESEARCH, 12(5), 260-280. https://doi.org/10.26437/arzmee72

Vancouver

Shriwas MK, Sindhi K, Thakre P. Deep Learning-Based Multilingual Smart Farming System for Crop Recommendation and Nutrient Monitoring. AFRICAN JOURNAL OF APPLIED RESEARCH. 2026;12(5):260-280. doi:10.26437/arzmee72

BibTeX
@article{Shriwas2026,
  author = {Shriwas, M. K. and Sindhi, K. and Thakre, P.},
  title = {Deep Learning-Based Multilingual Smart Farming System for Crop Recommendation and Nutrient Monitoring},
  journal = {AFRICAN JOURNAL OF APPLIED RESEARCH},
  year = {2026},
  volume = {12},
  number = {5},
  pages = {260--280},
  doi = {10.26437/arzmee72}
}
Kostenlos starten →