FarmSage:AI-Powered Crop Recommendations for Smarter Farming

• 

  report

• 

  result

• 

  result

• 

  dashboard

🚀 Elevator Pitch FarmSage is an AI-powered crop recommendation system that suggests the best crops to grow based on soil nutrients and climate conditions — enabling data-driven decisions and promoting sustainable agriculture.

💡 About the Project Inspiration Agriculture is the backbone of our economy, yet many farmers rely on traditional knowledge rather than data. We were inspired to create a smart, accessible, and scalable tool that can help farmers make better decisions, increase yields, and protect soil health.

What It Does FarmSage analyzes the following input features:

Nitrogen ((N))

Phosphorus ((P))

Potassium ((K))

Temperature

Humidity

pH

Rainfall

Using a machine learning model, it predicts the best crop to grow for the given conditions. Results are shown through a clean and intuitive web interface built with Streamlit.

🏗️ How We Built It Data Preprocessing

Cleaned and structured the dataset

Normalized features for compatibility

Visualized using Matplotlib and Seaborn

Model Training

Tested multiple algorithms: KNN, Decision Tree, Random Forest

Random Forest achieved highest accuracy

Evaluated using classification metrics

​

Web App Deployment

Used Streamlit for the UI

Integrated model for real-time predictions

Hosted on local server for demo

⚔️ Challenges We Faced Sourcing a reliable and diverse agricultural dataset

Preventing overfitting while maintaining accuracy

Designing a UX that’s usable for farmers with minimal tech background

Interpreting feature importance for decision transparency

🏆 Accomplishments We're Proud Of Achieved over 95% model accuracy

Built a working ML model + UI in under 48 hours

Created a scalable solution with real-world applications

Learned a lot about the intersection of AI and agriculture

📘 What We Learned Practical use of classification models in agriculture

Data cleaning and preprocessing best practices

Streamlit deployment and UI/UX design

Importance of domain-specific research

🌱 What's Next for FarmSage Integrate live weather data using APIs

Add geolocation-based crop zoning

Provide fertilizer and irrigation recommendations

Launch multilingual mobile version for accessibility

Collaborate with agricultural NGOs for real-world testing

🔧 Built With Python

Pandas

NumPy

Scikit-learn

Streamlit

Matplotlib

Seaborn

GitHub

🔗 Try It Out GitHub Repo:https://github.com/Thanishaudayakumar/Crop-recommendation-system-using-ML

Built With

• git
• jupyter
• matplotlib
• notebook
• numpy
• pandas
• python
• scikit-learn
• seaborn
• streamlit