The agricultural sector in India, despite its status as a leading producer globally, grapples with low farm productivity, resulting in diminished incomes for farmers. Addressing this challenge requires a strategic approach centeredaround increasing productivity, thereby enhancing farmer livelihoods. Crucially, farmers must be equipped with the knowledge of which crops are best suited to their specific plots of land to optimize yield potential. This entails consideration of various factors such as temperature, humidity, soil pH, rainfall patterns, and nutrient composition. However, many farmers lack access to this vital information, leaving them uncertain about which crops to cultivate for maximum yield and profit. Thus, the implementation of crop recommendation systems powered by machine learning algorithms presents a promising solution. By leveraging data on environmental conditions andsoilproperties,thesesystemscan accuratelypredictsuitablecropchoicesfor individual farms, empowering farmers to make informed decisions and ultimately improve their productivity and income levels.
Introduction
Problem Statement:
Indian farmers often choose crops based on intuition or short-term profits, rather than scientific data like soil suitability or market demand. These uninformed decisions can lead to financial losses and have been linked to the growing rate of farmer suicides. Given that agriculture contributes over 20% to India’s GVA, poor crop choices can also impact the broader economy.
Proposed Solution:
The project proposes an AI-powered crop recommendation system that leverages machine learning (ML) to guide farmers in selecting the most suitable crops based on:
Soil parameters: pH level, type, and nutrient concentration
Economic factors: production cost, market price, and yield data
The system aims to improve yield, profitability, and decision-making by providing data-driven crop suggestions.
Key Components:
1. Machine Learning Techniques:
The system uses an ensemble of ML algorithms for improved accuracy:
Random Forest: Uses multiple decision trees to reduce overfitting and improve prediction accuracy.
Decision Tree: Provides clear logic paths for decision-making based on feature splits.
Support Vector Machine (SVM): Handles complex, non-linear relationships and works well with large datasets.
Neural Networks: Implemented using Keras for pattern recognition and crop classification.
K-Nearest Neighbors (KNN): Suggests crops based on similar regional profiles.
Linear Regression: Used to evaluate the relationship between environmental conditions and crop production.
Implementation Details:
Data Sources: Kaggle, Indian government open data portals (data.gov.in)
Data Preprocessing: Includes cleaning, normalization, and filtering crops based on soil nutrient suitability.
Profit Analysis: Incorporates economic viability by calculating profit margins from historical data.
Web Application: A user interface for farmers to input their region, soil, and environmental data.
Performance & Evaluation:
Algorithm
Accuracy
Neural Network
89.88%
KNN with Cross Validation
88%
Linear Regression
88.26%
KNN
85%
Naive Bayes
82%
Decision Tree
81%
SVM
78%
Literature Review Highlights:
Prior works focused on fertilizer recommendations, android apps for nutrient management, and IoT-cloud systems for precision farming.
Most earlier models lacked integrated decision-making, combining both agronomic and economic factors.
This system stands out by merging real-time environmental data, historical performance, and profit analysis.
Conclusion
For the purposes of this project we have used three popular algorithms: Linear regression, Logistic regression and Neural network. All the algorithms are based on supervised learning.Our overall system is divided into three modules
1) Profitanalysis
2) Croprecommender
3) Crop Sustainabilitypredictor
This system assists farmers in selecting the optimal crop by offering insights that are typically overlooked, thereby reducing the risk of crop failure and enhancing productivity.It also helps in preventing losses. The system can be expanded to the web, making it accessibleto millions of farmers nationwide. We achieved an accuracy rateof 89.88% from the neural network and 88.26% from the linear regression model. Future enhancements involve integrating the crop recommendation system with a yield predictor subsystem, providing farmers with production estimates for the recommended crops.This solution offers farmers the opportunity to enhance agricultural productivity, mitigate soil degradation, anddecrease fertilizer usage by suggesting suitable crops based on diverse attributes. It provides a holistic prediction by factoring in geographical, environmental, and economic factors.
References
[1] S. P. Mohanty, U. Choppali and E. Kougianos, \"Everything You Wanted to Know About Smart Cities: The Internet of Things is the Backbone,\" IEEE Consumer Electronics Magazine, vol. 5, no. 3, pp. 60-70, July 2016.
doi: 10.1109/MCE.2016.2556879
[2] A. R. Patil and D. Patil, \"Crop Recommendation System Using Machine Learning Algorithms,\" 2020 International Conference on Communication and Signal Processing (ICCSP), Chennai, India, 2020, pp. 0878-0882.
doi: 10.1109/ICCSP48568.2020.9182108
[3] S. Sharma and V. P. Singh, \"Crop Recommendation System Using Machine Learning: A Review,\" 2021 5th International Conference on Intelligent Computing and Control Systems (ICICCS), Madurai, India, 2021, pp. 1006-1011.
doi: 10.1109/ICICCS51141.2021.9432247
[4] S. Chatterjee, A. Sarkar and M. Das, \"Crop Recommendation System Using Machine Learning for Agricultural Sector,\" 2021 4th International Conference on Intelligent Sustainable Systems (ICISS), Palladam, India, 2021, pp. 1526-1530.
doi: 10.1109/ICISS49785.2021.9616827
[5] K. A. Patil and N. R. Kale, \"A Model for Smart Agriculture Using IoT,\" 2016 International Conference on Global Trends in Signal Processing, Information Computing and Communication (ICGTSPICC), Jalgaon, India, 2016, pp. 543-545.
doi: 10.1109/ICGTSPICC.2016.7955300
[6] T. Singh, A. Yadav and R. Tripathi, \"A Crop Recommendation System Using Machine Learning,\" 2020 9th International Conference on System Modeling and Advancement in Research Trends (SMART), Moradabad, India, 2020, pp. 180-183.
doi: 10.1109/SMART50582.2020.9337127
[7] P. Patel and N. Patel, \"Crop Yield Prediction Using Machine Learning Techniques,\" 2021 International Conference on Emerging Smart Computing and Informatics (ESCI), Pune, India, 2021, pp. 245-249.
doi: 10.1109/ESCI50559.2021.9396910
[8] P. A. Adnan, M. Hasan, T. Islam, and M. M. Rahman, \"Machine Learning Based Soil Analysis and Crop Suggestion Using IoT,\" 2021 International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering (IC4ME2), Rajshahi, Bangladesh, 2021, pp. 1-4.
doi: 10.1109/IC4ME251882.2021.9380403
[9] B. P. Reddy and A. B. Nandimath, \"IoT and Machine Learning Based Crop Recommendation System,\" 2020 IEEE International Conference on Advances and Developments in Electrical and Electronics Engineering (ICADEE), Coimbatore, India, 2020, pp. 1-6.
doi: 10.1109/ICADEE49888.2020.9276850
[10] M. Singh, S. Rani and M. Kamal, \"Crop Recommendation System to Maximize Crop Yield Using Machine Learning Technique,\" 2019 International Conference on Smart Systems and Inventive Technology (ICSSIT), Tirunelveli, India, 2019, pp. 614-617.
doi: 10.1109/ICSSIT46314.2019.8987940
[11] R. Kumar and M. L. Garg, \"Data Mining Techniques in Crop Yield Prediction and Crop Suitability,\" 2018 2nd International Conference on Micro-Electronics and Telecommunication Engineering (ICMETE), Ghaziabad, India, 2018, pp. 255-258.
doi: 10.1109/ICMETE.2018.00062
[12] G. V. Kumar and B. Chatterjee, \"Smart Agriculture Using Machine Learning,\" 2021 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS), Greater Noida, India, 2021, pp. 738-742.
doi: 10.1109/ICCCIS51004.2021.9397120
[13] M. C. Sujatha, K. Vani, and V. M. Nandhini, \"Prediction of Crop Yield Using Regression and Decision Tree Algorithms,\" International Journal of Computer Sciences and Engineering, vol. 7, no. 6, pp. 239-243, 2019.
[14] A. Kale and S. R. Gengaje, \"Smart Farming: A Crop Recommendation System Using Machine Learning,\" International Research Journal of Engineering and Technology (IRJET), vol. 7, no. 6, pp. 2138-2143, June 2020.
[15] A. Sharma and A. K. Suryawanshi, \"IoT and Machine Learning in Agriculture: Crop Prediction Model,\" Journal of Emerging Technologies and Innovative Research, vol. 7, no. 6, pp. 56-61, June 2020.
[16] R. Singh, S. Saini and A. Girdhar, \"Machine Learning Based Crop Recommendation System Using Soil and Environmental Parameters,\" International Journal of Scientific & Engineering Research, vol. 10, no. 4, pp. 1446-1450, April 2019.
[17] \"Crop Recommendation Dataset,\" Kaggle, 2020. [Online]. Available: https://www.kaggle.com/datasets/atharvaingle/crop-recommendation-dataset
[18] \"Weather Data Archive,\" Indian Meteorological Department, 2021. [Online]. Available: https://mausam.imd.gov.in/
[19] P. Pandey, R. K. Mishra and D. P. Yadav, \"Smart Farming Using Decision Tree Algorithm,\" 2021 International Conference on Computer Communication and Informatics (ICCCI), Coimbatore, India, 2021, pp. 1-5.
doi: 10.1109/ICCCI50826.2021.9402432
[20] V. D. Jadhav and P. D. Patil, \"Crop Prediction System Using Machine Learning Algorithms,\" 2020 IEEE Pune Section International Conference (PuneCon), Pune, India, 2020, pp. 1-4.
doi: 10.1109/PuneCon50868.2020.9362364