Green Modern Bold Agriculture Presentation_20251012_152644_0000.pdf

Green Modern bold Agriculture presentation

1. MKisan Agricultural Portal Using
Machine Learning 20
25
presented by : Sakshi Chaudhary

2. Abstract Agriculture is the foundation of India’s economy and provides
employment to
Farmers face several challenges such as climate change,
unpredictable rainfall, improper fertilizer use, and limited
access to markets.
The MKisan Agricultural Portal is designed to address these challenges by
integrating Machine Learning (ML) and Data Analytics into one interactive
web platform.
The portal provides crop recommendation, fertilizer recommendation, yield
prediction, weather forecasting, agriculture news feed, and an online
trading platform.

3. The system analyzes soil nutrients, pH, temperature, rainfall, and
humidity using ML models to provide accurate recommendations.
It also allows farmers to sell crops directly to customers, eliminating
middlemen and ensuring better profits.
Thus, this project aims to bridge the gap between agriculture
andtechnology to promote smart farming and digital agriculture.

4. 2. ObjectivesTo design and develop a user-friendly web portal
for farmers using machine learning. To predict crop yield before cultivation to plan production
and sales. To recommend the most suitable crops for a
farmer’s land based on soil and weather data. To provide fertilizer recommendations for soil
health improvement. To facilitate direct trade between farmers and consumers
through the portal. To offer real-time agricultural news and weather updates

5. Farmers lack reliable tools to analyze soil, weather, and
market conditions.
They often make cultivation decisions without scientific or data
support.
Poor fertilizer management and unpredictable climate changes
reduce yield.
Lack of direct access to markets forces farmers to sell at
lower prices.
Existing portals provide limited features or focus on single
tasks like weather or market data.
Hence, there is a need for a comprehensive web-based
platform that integrates machine learning with agriculture to
assist farmers in all stages of crop production and sales.
3. Problem statement

6. 4. Scope of the
Project 1.The project focuses on developing a Decision Support System for Indian
farmers.
2.It uses datasets from Kaggle for:
CROP RECOMMENDATION
Fertilizer recommendation
Yield prediction
3.The project applies ML models such as Decision Tree, Random Forest, Support
Vector Machine, and Linear Regression. 4.The web portal provides three user modules:
Farmer: Access recommendations, predictions, weather updates, and trade options.
Customer: Buy crops directly from farmers.
Admin: Manage all users, monitor transactions, and maintain data.
5.The system ensures easy access, accurate recommendations, and secure transactions.

7. 5. Methodology 1.Data Collection:
Datasets were obtained from Kaggle related to soil, weather, and crop production.
2.Data Preprocessing:
Removal of null values and duplicates.
Encoding of categorical variables using label encoding.
Normalization of numerical features for better model accuracy.
3.Data Analysis:
Performed exploratory data analysis using visualization tools like Seaborn and
Matplotlib.
4.Model Training:
Applied machine learning algorithms such as:
Random Forest and Decision Tree for classification (recommendations)
Linear Regression and Random Forest Regressor for prediction tasks
Dataset split into 80% training and 20% testing.

8. A. FARMER
MODULE Sign up / login to the portal.
Access crop, fertilizer, and yield
prediction tools.
View news feed and weather
forecasts.
Sell crops online and manage crop
listings.B. CUSTOMER
MODULE
Create account and log in to
portal.
View available crop stocks.
Buy crops directly from farmers
and make online payments.
6. System Modules

9. 7. Front-End Design
3.Features:
Login/Registration Pages
Interactive Dashboard for Farmers and Admin
Data Input Forms for crop and fertilizer recommendation
Visualization pages for yield prediction results
Trade page for crop selling/buying
4.Advantages:
Simple navigation, multilingual capability (can be extended), and
accessibility on desktop and mobile.
1.Languages and Tools:
HTML5, CSS3, JavaScript, Bootstrap 5
2.Purpose:
To provide a clean, responsive, and user-friendly interface for farmers and
customers.

10. 8. Conclusion
1.The MKisan Agricultural Portal successfully applies Machine Learning to real-
world agricultural problems.
2.It empowers farmers by providing intelligent recommendations and yield
predictions.
3.The system’s trading and news modules promote transparency and awareness
among users.
4.By bridging the gap between agriculture and technology, this project
contributes to the vision of Digital India and Smart Agriculture.
5.With future IoT and mobile integration, this project can become a complete
digital ecosystem for sustainable farming.

11. Rimberio Co 9. Results and Performance Random Forest algorithm achieved
~95% accuracy for crop
recommendation.

12. Thank You