Republic of the Philippines

Davao Oriental State University

A university of excellence, innovation and inclusion
FACULTY OF COMPUTING, ENGINEERING AND
TECHNOLOGY
Guang-guang, Dahican, City of Mati, Davao Oriental 8200
Tel.: +63 (087) 3883 195
Website: www.dorsu.edu.ph

Proposed Title of Research Interest

Web-Based AI-Driven Agricultural Market Price Prediction System for Davao Oriental

Area of Study

Business Analytics & Market Forecasting, Machine Learning & Predictive Analytics, Web-Based AI Applications,
Agricultural Data Science & AgriTech, Cloud-Based AI & Web Development

Introduction

Agriculture is the backbone of Davao Oriental's economy, with many farmers relying on the cultivation of rice,
coconut, banana, corn, and other high-value crops. However, market price volatility remains a significant
challenge, as agricultural prices fluctuate due to climatic conditions, supply-demand imbalances, global trade
policies, and economic factors. Farmers and traders often struggle to predict price movements, leading to financial
losses, inefficient resource allocation, and missed opportunities for profit optimization.

Traditional forecasting methods in Davao Oriental rely on historical price trends, manual estimates, and
middlemen negotiations, which are often unreliable and lack real-time insights. While some digital tools exist, they
are complex, costly, and not specifically tailored to local agricultural conditions.

This study proposes the development of a Web-Based AI-Driven Agricultural Market Price Prediction System,
designed to empower farmers, traders, and policymakers in Davao Oriental with real-time, AI-powered market
price forecasts. The system will leverage machine learning, big data analytics, and cloud computing to analyze
historical trends, weather conditions, economic indicators, and supply-demand data, providing accurate and
actionable price predictions.

This project is highly relevant to the field of Information Technology (IT) as it demonstrates how AI, predictive
analytics, and web-based technologies can be applied to solve real-world economic challenges. By developing an
accessible and data-driven decision-support tool, this study contributes to the digital transformation of agriculture,
helping stakeholders make smarter trading and financial decisions in an increasingly data-driven world.

Problem Statement

Agriculture is a key driver of economic activity in Davao Oriental, with many farmers depending on crops such as
rice, coconut, banana, corn, and other high-value commodities. However, the unpredictability of agricultural
market prices remains a critical challenge, as farmers, traders, and agribusinesses struggle to anticipate price
fluctuations and market trends. Agricultural prices are influenced by various external factors, including weather
conditions, supply chain disruptions, economic policies, inflation rates, transportation costs, and global trade
dynamics. Without access to accurate price forecasting tools, farmers are often forced to sell their produce at low
prices, resulting in financial losses, reduced profitability, and economic instability. This problem is further
exacerbated by the seasonality of crops, where prices can swing drastically due to overproduction, natural
calamities, or sudden demand surges.

Currently, farmers in Davao Oriental rely on traditional market prediction methods, such as word-of-mouth
information, middlemen estimate, and historical price trends, which are often outdated and unreliable. While
some government agencies provide periodic price updates, these reports lack real-time insights and do not factor
in sudden changes in market conditions. Additionally, many existing agricultural intelligence platforms are too
expensive, complex, or designed for large-scale commercial agribusinesses, making them inaccessible to
smallholder farmers and local traders. As a result, farmers struggle to make data-driven decisions, leading to issues
such as post-harvest losses, inefficient crop planning, and missed opportunities for better market timing.

Moreover, climate change and environmental factors have introduced new uncertainties in crop production,
affecting supply chains and making price movements even more difficult to predict. Sudden weather changes,
typhoons, droughts, and pest infestations can drastically impact harvest yields, causing supply-demand imbalances
that further contribute to price volatility. In addition, economic factors such as rising fuel prices, labor shortages,
and shifting government trade policies also play a crucial role in agricultural pricing, making manual price
forecasting methods inadequate.

Given these challenges, there is a pressing need for an AI-driven, web-based solution that can analyze historical
data, monitor real-time market conditions, and generate accurate price forecasts to support farmers, traders, and
policymakers in Davao Oriental. This study proposes the development of a Web-Based AI-Driven Agricultural
Market Price Prediction System, which will leverage machine learning algorithms, big data analytics, and real-time
API integrations to provide accurate, data-driven price predictions for key agricultural commodities. By
incorporating economic indicators, weather patterns, and historical price trends, this system will offer valuable
market insights, enabling farmers and traders to optimize their selling strategies, reduce financial risks, and
improve income stability. Additionally, this research aims to bridge the gap between technology and agriculture,
making predictive analytics accessible, user-friendly, and beneficial to rural farming communities. By providing
real-time and accurate market price forecasts, this system will contribute to more efficient trading, better resource
management, and greater economic resilience in the agricultural sector of Davao Oriental.

Objectives of the Study

General Objective:

The primary objective of this study is to develop and implement a Web-Based AI-Driven Agricultural Market Price
Prediction System for Davao Oriental, enabling farmers and traders to make informed, data-driven decisions by
providing real-time and accurate crop price forecasts.

Specific Objectives:

1. To design and develop a web-based AI system for predicting agricultural market prices using machine
learning models.
2. To integrate real-time external data sources such as weather patterns, inflation rates, and trade policies
into the forecasting model.
3. To evaluate the system’s performance using Mean Absolute Error (MAE), Mean Squared Error (MSE), and
R² Score.
4. To create a user-friendly, mobile-responsive platform accessible to farmers and traders.
5. To assess the usability and effectiveness of the system through feedback from agricultural stakeholders.

Research Locale and Respondents

Research Locale:

The study will be conducted in Davao Oriental, focusing on various farming communities, local markets, and
agricultural trade hubs where price fluctuations significantly impact stakeholders.

Target Respondents:

 Farmers and agricultural producers

 Traders and wholesalers

 Government agencies and agricultural offices

 Agricultural cooperatives and agribusiness owners

 Researchers and agricultural economists

Scope and Delimitations:

This study focuses on the development and implementation of a Web-Based AI-Driven Agricultural Market Price
Prediction System designed to assist farmers, traders, and policymakers in Davao Oriental by providing real-time,
data-driven crop price forecasts. The system will utilize machine learning algorithms, big data analytics, and real-
time economic and weather data to generate accurate price predictions for selected agricultural commodities. It
will be developed as a web-based platform, ensuring accessibility through computers, tablets, and mobile devices.

Scope:
 Market Price Prediction: The system will analyze historical price trends, weather conditions, supply-
demand data, and economic indicators to forecast agricultural market prices in Davao Oriental.
 Machine Learning Integration: AI models such as XGBoost, LSTM (Long Short-Term Memory), and ARIMA
 (AutoRegressive Integrated Moving Average) will be implemented to improve the accuracy of price
forecasts.
 Real-Time Data Sources: The system will integrate weather APIs, government agricultural reports, and
economic data sources to enhance forecasting precision.
 User Accessibility: The platform will be designed to be mobile-responsive and user-friendly, ensuring
farmers, traders, and policymakers can easily access and interpret market insights.
 Data Visualization: The system will feature interactive dashboards, trend analysis graphs, and prediction
models, providing clear and actionable insights for stakeholders.
 Evaluation Metrics: The system's prediction accuracy will be measured using Mean Absolute Error (MAE),
Mean Squared Error (MSE), and R² Score, ensuring reliability.

Delimitations:
 Geographical Limitation: The study will focus only on agricultural price trends in Davao Oriental, meaning
the system may not accurately predict prices in other provinces or international markets.
 Limited Crop Selection: The system will initially be trained on a limited number of high-value crops (e.g.,
rice, coconut, banana, and corn) based on available data and economic impact.
 Internet Dependency: Since the platform relies on real-time data updates, users will need an active
internet connection to access predictions.
 Non-Financial Market Predictions: The system will not cover other financial aspects such as loan
forecasting, investment planning, or farm subsidies, as it is specifically focused on crop price prediction.
 Limited User Input Influence: While the system provides data-driven predictions, external factors (e.g.,
sudden policy changes, export restrictions, or unexpected global crises) may still cause unforeseen price
fluctuations that the AI model cannot fully anticipate.
Methodology

This study will follow the Agile development methodology, a flexible and iterative approach that allows for
continuous improvements based on user feedback and real-world testing. This methodology is ideal for developing
the Web-Based AI-Driven Agricultural Market Price Prediction System, as it ensures that the system evolves to
meet the specific needs of farmers, traders, and policymakers in Davao Oriental. The Agile approach will enable
incremental development, testing, and refinements, ensuring that the system is user-friendly, accurate, and
scalable.

The project will be developed using a combination of machine learning, web technologies, and cloud computing to
facilitate real-time price forecasting and data-driven decision-making. The backend will be built using Python with
Flask or FastAPI, enabling efficient handling of AI-based predictions and API integrations. The frontend will be
designed using React.js or Vue.js, ensuring a responsive and interactive user experience accessible across devices,
including mobile phones and tablets. The database will be managed using PostgreSQL or MySQL, allowing for
secure and structured storage of historical crop prices, weather conditions, and supply-demand data. The system’s
core functionality will rely on machine learning models, including XGBoost, LSTM (Long Short-Term Memory), and
ARIMA (AutoRegressive Integrated Moving Average). These models will be trained using historical price trends,
meteorological conditions, and economic indicators, enabling accurate price forecasting for crops commonly
grown in Davao Oriental. Additionally, real-time data sources, such as weather APIs, government market reports,
and trade data from FAO (Food and Agriculture Organization), will be integrated to enhance prediction accuracy.

The development process will be structured into six major phases to ensure systematic execution. The first phase,
Planning & Requirements Gathering, will involve conducting interviews, surveys, and research to identify the
specific challenges farmers and traders face regarding market price volatility. This phase will also involve data
collection from local agricultural offices, government agencies, and online repositories to build a robust dataset for
machine learning training. The second phase, System Design, will focus on defining the system architecture,
designing UI/UX wireframes, and planning API integrations for real-time data updates. During the Development &
Implementation phase, the machine learning models will be trained using historical and real-time data, and the
web application will be developed to provide an interactive dashboard displaying market trends, price forecasts,
and weather conditions. The integration of real-time APIs for meteorological and economic data will ensure the
system remains dynamic and up-to-date.

Following development, the Testing & Evaluation phase will assess the accuracy, reliability, and usability of the
system. The machine learning models will be evaluated using industry-standard metrics such as Mean Absolute
Error (MAE), Mean Squared Error (MSE), and R² Score to determine the precision of price forecasts. Additionally,
usability testing will be conducted with a group of farmers, traders, and agricultural policymakers, who will provide
feedback on the system’s accessibility, ease of use, and effectiveness in real-world decision-making. Based on this
feedback, necessary refinements will be made before deployment.

The Deployment & Monitoring phase will focus on making the system publicly accessible by hosting it on Google
Cloud, AWS, or Heroku. This will ensure that users in Davao Oriental can access the tool anytime and from any
device with an internet connection. Security measures will also be implemented to protect user data and API
connections from unauthorized access. Finally, the Maintenance & Enhancement phase will ensure the system
remains functional, accurate, and continuously improves over time. New datasets will be incorporated periodically,
allowing the AI models to learn from recent trends and improve predictions. Future enhancements may include
mobile app development for offline access and blockchain technology integration to enhance transparency in the
agricultural supply chain.

By following this structured methodology, the Web-Based AI-Driven Agricultural Market Price Prediction System
will be developed as a highly functional, data-driven, and user-friendly tool that empowers agricultural
stakeholders in Davao Oriental with AI-powered market insights. This system will help reduce financial risks,
optimize trade decisions, and enhance economic stability in the province’s agricultural sector, ultimately
contributing to technological advancement in agriculture.

Expected Output

The expected outcome of this study is the successful development and deployment of a Web-Based AI-Driven
Agricultural Market Price Prediction System that provides real-time and accurate crop price forecasts to farmers,
traders, and policymakers in Davao Oriental. The system will serve as a decision-support tool, helping users make
informed choices on when and where to sell their produce by analyzing historical trends, weather conditions,
economic indicators, and supply-demand patterns.

The system will have the following key deliverables:

 A fully functional web-based platform that enables farmers and traders to access real-time price forecasts
through an interactive dashboard.
 A machine learning-based price prediction model capable of analyzing historical price data, weather
conditions, and economic factors to generate accurate market price forecasts.
 Integration with real-time data sources, such as weather APIs, government agricultural reports, and trade
databases, to ensure continuously updated and accurate predictions.
 A user-friendly and mobile-responsive interface that allows non-technical users to easily navigate and
interpret market trends, ensuring accessibility even for farmers in rural areas.
 A recommendation system that suggests optimal selling times and locations based on predicted price
trends, helping farmers maximize their profits and reduce losses from price volatility.
 A database system for storing historical market data, enabling further analysis and long-term trend
identification.
 A performance evaluation report that assesses the system’s accuracy using industry-standard metrics
such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and R² Score, ensuring the reliability of
price predictions.
 User feedback and usability report based on testing conducted with farmers, traders, and agricultural
stakeholders, evaluating the effectiveness, accuracy, and ease of use of the system.

Significance of the Study

This study is significant as it introduces a Web-Based AI-Driven Agricultural Market Price Prediction System,
designed to provide real-time, data-driven insights to farmers, traders, and policymakers in Davao Oriental. By
leveraging machine learning, predictive analytics, and cloud-based technologies, the system aims to reduce
financial risks, optimize decision-making, and improve agricultural market efficiency. The project holds immense
value for agriculture, business, and the field of information technology (IT) by demonstrating how AI-driven
solutions can enhance traditional market forecasting methods.

Farmers and Agricultural Producers:

This study will significantly benefit farmers by providing instant, AI-driven market price forecasts, enabling them to
strategically plan when and where to sell their crops for maximum profit. By integrating real-time weather and
economic data, the system will help reduce financial losses caused by sudden price drops and market fluctuations.
Additionally, access to data-driven insights will allow farmers to anticipate risks, make informed decisions, and
enhance overall sustainability in agricultural production.

Traders, Wholesalers, and Agribusiness Owners:

Agricultural traders and wholesalers will gain valuable market insights that allow them to optimize inventory
management and purchasing strategies. With real-time price forecasts, they can mitigate risks associated with
price volatility, ensuring stable trade and fair pricing negotiations with farmers. This system will contribute to a
more balanced and efficient agricultural supply chain, benefiting both producers and distributors.

Government Agencies and Policymakers:

This study will assist government agencies and agricultural policymakers by providing market intelligence that can
support the development of better trade regulations, subsidy programs, and pricing policies. By analyzing market
trends and price fluctuations, policymakers can proactively address supply shortages or surpluses, ensuring food
security and economic stability in Davao Oriental. The system will also help in monitoring agricultural market
trends to support evidence-based decision-making in the sector.

Future Researchers:
This study will contribute to the growing body of research on AI-driven market forecasting, serving as a valuable
reference for future studies in agricultural data science and predictive analytics. Future researchers can expand
upon this work by improving machine learning models, integrating additional economic variables, and developing
more localized forecasting systems. Additionally, this research will open opportunities for further studies on AI
applications in precision farming, agri-finance, and rural economic development.

The IT Community:
The project will demonstrate the practical application of AI, machine learning, and web-based predictive analytics
in agriculture. It will provide insights into the development and deployment of AI-powered decision-support
systems, serving as a reference for IT professionals and developers interested in building similar intelligent
forecasting platforms for other industries. Additionally, the study will highlight best practices in cloud-based AI
integration, big data analytics, and interactive web development, contributing to advancements in IT solutions for
business intelligence and economic forecasting.

References

Abiodun, O. I., Jantan, A., Omolara, A. E., Dada, K. V., Umar, A. M., & Linus, O. U. (2018). A review of machine
learning models for predicting agricultural prices. Artificial Intelligence Review, 52(1), 1-35.
https://doi.org/10.1007/s10462-018-9654-6

Baffes, J., & Haniotis, T. (2016). What explains agricultural price movements? Journal of Agricultural Economics,
67(3), 706-721. https://doi.org/10.1111/1477-9552.12169

Chandio, A. A., Jiang, Y., Rehman, A., & Gokmenoglu, K. K. (2020). Predicting the price volatility of agricultural
commodities using machine learning algorithms. Computers and Electronics in Agriculture, 169, 105200.
https://doi.org/10.1016/j.compag.2019.105200

FAO (Food and Agriculture Organization). (2021). Agricultural Market Information System (AMIS) Report. Retrieved
from https://www.fao.org/

Gouel, C. (2020). Agricultural price instability: A survey of competing explanations and remedies. Food Policy, 92,
101845. https://doi.org/10.1016/j.foodpol.2019.101845

Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: Principles and practice. OTexts. Retrieved from
https://otexts.com/fpp2/

Kumar, M., & Kalita, M. K. (2021). A machine learning-based approach for agricultural commodity price prediction.
International Journal of Data Science and Analytics, 11(2), 155-167. https://doi.org/10.1007/s41060-020-00218-7

Mishra, N., & Singh, B. (2022). Impact of climatic factors on agricultural price volatility: A machine learning
approach. Environmental Research, 204, 111995. https://doi.org/10.1016/j.envres.2021.111995

Shah, M., & Verma, A. (2019). AI-driven solutions for agricultural price forecasting: A review of recent
advancements. Journal of Computational Intelligence in Agriculture, 7(4), 210-225.
https://doi.org/10.1016/j.jcia.2019.04.012

World Bank. (2022). Agricultural price trends and food security analysis. Retrieved from
https://data.worldbank.org/