Farmer Support Website: Crop Recommendation Model

Abstract

The Farmer Support Website is a smart, multilingual platform designed to assist Indian farmers in making informed agriculturaldecisions. It integratesmachine learning models, real-time weather APIs, and community features to provide crop recommendations, fertilizer guidance, price predictions, and access to government schemes. The system uses a Random Forest classifier to suggest optimal crops based on soil type, and Linear Regression models trained on historical market datato estimate crop prices. A dedicated forum allows farmersto postquestionsandreceive insights14 from peers and experts in English, Hindi, or Marathi. The platform also features Firebase authentication, ensuring secure access and user-specific interactions. By combining data-driven insights with localized accessibility, the website aims to bridge the gap between modern agricultural technology and grassroots-level farmers.

Introduction

Introduction:
Agriculture is a crucial sector in India, but many farmers struggle with limited access to timely information, modern farming practices, and market data—often worsened by language barriers and digital illiteracy. The Farmer Support Website was developed as an integrated, multilingual platform that uses machine learning, real-time APIs, and community support to deliver actionable insights to Indian farmers.

Problem Statement:

Indian farmers face several challenges:

• No personalized crop or fertilizer advice.

• Inability to predict crop prices.

• Lack of localized weather updates.

• Poor awareness of government schemes.

• Language and digital literacy barriers.

• No central forum for expert or peer consultation.

Objectives:

1. Recommend crops and fertilizers based on soil using a Random Forest model.

2. Provide real-time weather updates via location-based APIs.

3. Predict crop prices using Linear Regression on historical data.

4. Support multilingual access (English, Hindi, Marathi).

5. Host a Farmer Forum for peer and expert Q&A.

6. Secure login and registration via Firebase Authentication.

7. Ensure a mobile-friendly, responsive design for rural accessibility.

Methodology Highlights:

• Frontend: HTML, Tailwind CSS, JS, with dynamic language switching.

• Authentication: Firebase for user login and registration.

• Crop/Fertilizer Recommendation: Random Forest trained on Maharashtra-specific soil-crop data.

• Price Prediction: Linear Regression trained on data from 2018–2024 for key crops.

• Weather Integration: Real-time weather via Visual Crossing API using geolocation.

• Farmer Forum: Lightweight chat system with real-time updates and future scope for moderation and expert replies.

• Multilingual Support: Platform and forum content available in English, Hindi, and Marathi.

Results:

• Crop Recommendation: Achieved 92% accuracy in predicting suitable crop-fertilizer combinations.

• Price Prediction: Less than ?100 mean absolute error for common crops like onion, tomato, and potato.

• Weather Data: Location-based, real-time forecasts integrated into decision tools.

• Multilingual UI: Fully functional in 3 languages, improving reach and usability.

• Forum System: Real-time peer interaction enabled with Firebase backend.

• Responsiveness: Optimized for low-end devices and rural internet conditions.

Discussion:

The platform effectively tackles major gaps in Indian agriculture by offering localized, data-driven, and user-friendly solutions. Its use of machine learning enhances crop planning, while real-time weather and price forecasts assist in practical decision-making. The multilingual interface helps overcome the digital divide. Though promising in tests, real-world success depends on internet access, farmer adoption, and institutional support.

Conclusion

The Farmer Support Website offers a practical and scalable solution for empowering farmers through intelligent, multilingual, and data-driven tools. By combining machine learning models, weather APIs, community interaction, and language localization, the platform addresses critical gaps in traditional agricultural advisory systems. The implementation of crop and fertilizer recommendation using a Random Forest classifier ensures personalized advice based on soil conditions. The price prediction feature gives farmers foresight into seasonal market trends, supportingbetter planning and storage decisions. Multilingual supportand a responsive design enable wider access across regional and linguistic boundaries. The project emphasizes that technology adoption inagriculture must go beyond functionality—it must be accessible, inclusive, and localized to be truly impactful.While the current system is functional and modular, it offers immense scope for integration with mobile apps, government databases, and satellite or IoT-based data for even moreprecise farming guidance. In conclusion, the Farmer Support Website stands as a promisingsteptowardsmarterandmoreconnectedagriculture in India.

References

[1] Breiman, L. (2001). \"Random forests.\" Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324 [2] Brownlee, J. (2017). \"A Gentle Introduction to Linear Regression for Machine Learning.\" Machine Learning Mastery. https://machinelearningmastery.com/linear-regression-for-machine-learning/ [3] Dutta, R., & Bose, S. (2020). \"Crop Recommendation System for Precision Agriculture.\" Procedia ComputerScience, 167,2401–2410. https://doi.org/10.1016/j.procs.2020.03.296 [4] Science Conference, 51–56. https://conference.scipy.org/proceedings/scipy2010/pdfs/mckinney.pdf [5] Ministry of Agriculture & Farmers Welfare. (2023). \"Government Schemes and Farmer Welfare Programs.\" Government of India. https://agricoop.nic.in [6] Pandas. (n.d.). \"Pandas Library Documentation.\" https://pandas.pydata.org/ [7] Patil, A. P., & Birajdar, S. (2019). \"Real-Time Crop Prediction System Based on Soil Parameters and Weather Conditions.\" International Journal of Scientific & Engineering Research, 10(5), 1230–1234. [8] Pawar, S., &Shelke, R. (2021). \"A Web-based Crop Recommendation System Using Decision Tree Algorithm.\" In 2021 2nd International Conference for Emerging Technology (INCET), 1–5. https://doi.org/10.1109/INCET51464.2021.9456213 [9] Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., et al. (2011). \"Scikit-learn: Machine learning in Python.\" Journal of Machine Learning Research, 12,2825–2830. https://jmlr.org/papers/v12/pedregosa11a.html [10] Visual Crossing Corporation. (n.d.). \"Visual Crossing Weather API Documentation.\" https://www.visualcrossing.com/weather-api [11] Waghmare, A. K., & Gawande, U. (2022). \"A Review on Agriculture Price Prediction Using Machine Learning Algorithms.\" InternationalJournal of Engineering Research & Technology (IJERT), 11(4), 528–532. [12] Firebase. (n.d.). \"Firebase Authentication Documentation.\" Google Developers. https://firebase.google.com/docs/auth [13] Government of Maharashtra. (2023). \"Soil and Crop Guidelines – Department of Agriculture.\" https://krishi.maharashtra.gov.in [14] Jain, R., Gupta, A., & Sharma, V. (2021). \"Smart Agriculture Using IoT, Machine Learning and Cloud Computing.\" Materials Today: Proceedings, 47(1), 400–406. https://doi.org/10.1016/j.matpr.2021.06.086 [15] Kumar, N., & Singh, A. (2020). \"A Survey on Machine Learning Techniques in Agriculture.\" In 2020 5thInternational Conference on Communication and Electronics Systems (ICCES),381–386.https://doi.org/10.1109/ICCES48766.2020.9137902 [16] McKinney,W.(2010).\"DataStructuresforStatistical ComputinginPython.\"Proceedingsofthe9thPythonin

Copyright

Copyright © 2025 Deepika Pede, Karuna Kanthi, Shubham Vernekar , Siddhi Tiwari , Shantanu Kirpane . This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.