Abstract
The abstract highlights the importance of agriculture in providing food for people and raw materials for industries. It introduces a smart irrigation system that predicts water requirements for crops using machine learning algorithms based on parameters like moisture, temperature, and humidity. This system aims to improve crop yield, reduce irrigation wastage, and assist farmers in decision-making by sending alerts regarding water supply needs.
Introduction
The introduction highlights the role of Internet of Things (IoT) technology in monitoring and controlling devices globally, connecting them with living entities. IoT advancements have improved various aspects of life, such as education, healthcare, and automation, with a focus on addressing fundamental needs like food production from agricultural fields. The text emphasizes the importance of IoT in enhancing agricultural practices to meet the increasing food demand due to population growth and climate challenges.
Previous works
In the related work section, the text discusses various research studies focusing on smart irrigation systems using IoT and machine learning techniques to optimize water usage in agriculture. For instance, in reference [1], an irrigation system is proposed that leverages machine learning to analyze atmospheric and soil data for efficient water management in crop fields, providing real-time decision-making capabilities to farmers through mobile and web applications. These technologies aim to reduce water wastage, improve crop productivity, and offer personalized irrigation solutions based on plant types and environmental conditions.
Method proposal
The proposed method in the text refers to the use of Supervised Machine Learning algorithms, where labeled data with known answers is provided to the machine to learn patterns and solve problems. This process involves training the machine on the data to understand patterns and then testing it to predict suitable answers based on the learned patterns. The accuracy of the results depends on factors like the size of the data, the algorithms used, and the presence of noise or outliers in the input data.
About algorithm
The decision tree algorithm is a simple and efficient classification algorithm used in supervised learning to solve regression and classification problems. It aims to train a model that can predict the value or class of a target variable by generating clear decision rules based on training data. The algorithm involves starting at the root node and following paths through decision nodes to reach leaf nodes, making predictions based on attribute validations along the way.
About architecture
The architecture of the proposed system includes components such as temperature, soil moisture, humidity sensors, and a Raspberry Pi. Raspberry Pi serves as the central component, storing datasets, hosting a web server, and receiving data from sensors for processing using the decision tree algorithm to predict outcomes related to water supply for farmers. Data collected from sensors is stored in a cloud database for future use.
Experiment results
In the results section, the output of the system is presented, showing values for temperature in both Centigrade and Fahrenheit, humidity, water presence, and alerts sent to farmers via email based on the decision tree algorithm's output on whether to water the crops. The system stores data on temperature, humidity, and water presence in cloud storage for future reference, aiming to help farmers make informed decisions on watering crops efficiently to address water scarcity concerns in agriculture.
Conclusion of article
The conclusion of the study emphasizes the importance of addressing future water scarcity in agriculture by implementing smart technology to optimize water usage. The proposed model, utilizing a decision tree algorithm trained on real-time sensor data, enables efficient irrigation practices by providing timely alerts to farmers via email, helping them make informed decisions on watering crops and reducing water wastage. The research aims to enhance land productivity and sustainability in agriculture through the integration of technology and data-driven decision-making.