Farmer Decision Support System: A Data-Driven

Approach to Agricultural Optimization

Nikunj Mittal, Poornachandra K S, Nagendra Saketh Kashyap
Department of Computer Science and Engineering
R.V. College of Engineering, Bengaluru, India
Email: {1RV22CS129, 1RV22CS138, 1RV22CS124}@rvce.edu.in

Abstract—The Farmer Decision Support System (FDSS) is

designed to empower farmers with personalized, data-driven make informed decisions about selling their produce at
insights to boost agricultural productivity and profitability. By the right time for maximum profit.
analyzing historical weather patterns, soil health metrics, and • Weather Data Integration: Fetches real-time weather
market trends, the system generates tailored recommendations updates and forecasts to assist farmers in planning plant-
for crop selection, planting schedules, and fertilizer use. The
ing and harvesting schedules while mitigating risks from
system integrates real-time data sources and predictive modeling
techniques to enhance decision-making and minimize risks. unpredictable climatic changes.
Leveraging machine learning techniques such as gradient boost-
III. IMPLEMENTATION
ing and logistic regression, FDSS delivers precise, real-time
advice while ensuring accessibility for rural farmers. The FDSS A. Machine Learning Models
also emphasizes usability, offering an intuitive interface
accessible via mobile and web platforms, thereby bridging the The FDSS employs various machine learning models for
technological gap in rural communities. different predictive tasks:
Index Terms—Agriculture, Decision Support System, Machine • Gradient Boosting: Used for crop yield predictions by
Learning, Data Analytics, Crop Prediction, Precision Farming learning from past data and improving prediction accu-
racy over time.
I. INTRODUCTION • Logistic Regression: Applied to classify optimal crop
Traditional farming methods rely heavily on experience selections based on input features such as soil quality,
and intuition, often leading to inefficiencies in crop selection, climate conditions, and seasonal factors.
resource utilization, and financial planning. Rapid changes • Random Forest: Helps in identifying the best fertilizer
in climate, soil conditions, and market fluctuations further recommendations by analyzing nutrient requirements and
complicate the decision-making process for farmers. Recent soil deficiencies.
advancements in machine learning and data analytics provide The models are trained using diverse datasets that encom-
an opportunity to bridge this gap by offering data-driven pass regional weather patterns, soil compositions, and crop
recommendations that enhance productivity and sustainability. performance records. This ensures that FDSS can generalize
The Farmer Decision Support System (FDSS) aims to its predictions across different agricultural landscapes.
integrate historical weather data, soil parameters, and market
trends to assist farmers in making informed decisions. The B. User Interface and Accessibility
system’s key objectives include reducing financial risks, The FDSS is designed with a user-friendly interface to cater
optimizing agricultural resources, and improving crop yields to farmers with varying levels of digital literacy. It supports
by leveraging technology. both web and mobile platforms, providing:
• Graphical insights and visual charts for better data inter-
II. SYSTEM ARCHITECTURE pretation.
The FDSS consists of the following core components: • Offline functionality to enable access in remote areas

• User Management Module: Allows farmers to register, with limited internet connectivity.
log in, and manage their preferences. This ensures a per- • Interactive dashboards that allow users to easily navigate

sonalized experience tailored to their specific agricultural through recommendations and insights.
needs. • Customizable alerts and notifications for timely agricul-

• Crop Prediction Module: Analyzes weather conditions, tural actions.

soil properties, and historical data to recommend the best IV. RESULTS AND DISCUSSION
crops for cultivation. It also considers disease resistance
Initial testing of FDSS showed an accuracy of approxi-
and water requirements to provide optimal recommenda-
mately 88.89% in crop yield predictions, outperforming tradi-
tions.
tional heuristic-based methods. The system was tested across
• Market Analysis Module: Provides insights into histor-
multiple regions, demonstrating high adaptability to different
ical and real-time crop price trends, enabling farmers to
climatic and soil conditions. Farmers using the system
reported
improved decision-making capabilities, optimized fertilizer
usage, and increased profitability. The user feedback high-
lighted the benefits of real-time insights and personalized
recommendations, making FDSS a valuable tool for modern
agriculture.
V. CONCLUSION AND FUTURE ENHANCEMENTS
The FDSS serves as a robust decision support system for
farmers, integrating data analytics with agricultural practices.
By combining machine learning models with real-time envi-
ronmental data, the system enhances farming efficiency and
promotes sustainable agricultural practices. Future enhance-
ments include:
• Pest and Disease Prediction: Implementing deep learn-
ing models to detect early signs of crop diseases and pest
infestations.
• IoT Sensor Integration: Connecting IoT-based soil and
climate sensors to provide real-time data collection and
analysis.
• Blockchain for Supply Chain Transparency: Ensuring
fair pricing and traceability in the agricultural supply
chain.
• Automated Irrigation Recommendations: Optimizing
water usage through AI-driven irrigation strategies.
ACKNOWLEDGMENT
We extend our gratitude to Dr. Suma B, Assistant
Professor, RVCE, for her invaluable guidance throughout the
project. We also thank our peers and mentors for their support
and constructive feedback during the development of this
system.
REFERENCES
[1] S. Timotheou et al., ”Beamforming for MISO interference channels with
QoS and RF energy transfer,” IEEE Trans. Wireless Commun., vol. 13,
no. 5, pp. 2646–2658, May 2014.
[2] M. Young, The Technical Writer’s Handbook. Mill Valley, CA: Univer-
sity Science, 1989.
[3] R. Kumar et al., ”A Decision Support System for Crop
Recommendation Using Machine Learning Classification Algorithms,”
Res. Gate, 2023.
[4] Meteomatics AG, ”Meteomatics Weather API Documentation,” 2024.
Available: https://www.meteomatics.com/en/api/