Machine Learning Project

Machine Learning

What is Machine Learning (ML)? is a branch of artificial intelligence that focuses on developing techniques wherin computers can learn form data and perform spesific task without being explicitly programmed.

The project I am working on involves the development of ontology using the CNN-LSTM method. In the diagram, it can be seen that I am trying to modify or implement the LSTM model on a sentence input, converting it into 2-dimensional data so that the system can make predictions.

What Do you make?

I am creating an ontology-based prediction system using the CNN-LSTM method. Why use ontology? There are several reasons why ontology is used as the foundation for building this system, including :

• Structured Knowledge Representation
  Ontology allows for the structured Representation of knowledge about pest, diseases, sympyoms, and their relationships. This make the information more understandable and accesable for computer systems.
• Data Interoperability
  Ontology supports data interoperability, enabling the integration of data from various sources. By using standard ontology, data can be combined and analyzed efficiently.
• Inference Capability
  Ontology-based systems can use inference engines to generate new knowledge from existing knowledge. This allows for smarter diagnostics and more accurate predictions about pests and diseases.
• Flexibility and Scalability
  Ontologies can be easily updated and expanded by adding new concepts or relationships without disrupting the existing structure. This is important for keeping up with new developments in agricultural knowledge.
• Contextual Understanding
  Ontologies enable better contextual understanding. The system can comprehend how various symptoms and environmental conditions are related, providing more accurate diagnoses.
• Decision Support
  Using ontology, the system can offer better decision support for farmers and agricultural experts. The system can provide recommendations for appropriate actions based on the diagnosis.
• Reduced Dependence on Experts
  Ontology-based systems can reduce dependence on experts by automating the diagnostic process. This is particularly important in areas with a shortage of experts.

These reasons support my use of ontology as the foundation for building the desired system.

What was the input and output?

The input for the system is textual. In this project, I am constructing data using ontology. Focusing on the discussion topic, the output provided by the system I developed includes the identification of the disease or pest affecting the maize plant, methods to control the pest and disease, and actions to take when the plant is affected by the disease.

Any Advice?

From the system I developed, there are recommendations from several experts. These experts are individuals from a group of maize plant specialists. They stated, "If this research has a broader scope, it will be beneficial to many people, especially to plant enthusiasts."