Diagnosis of Long Sightedness Using Neural Network and Decision Tree Algorithms

Abstract

Long-sightedness occurs provided the eyes don't concentrate properly on the retina that is the delicate illuminated portion at the back of the eyes. It influences the capability to recognise nearby items. Numerous researches have been conducted on eye disease, but the outcome hasn't been 100% accurate. What motivated the development of this system was that most people are not aware of the warning signs of the disease as a result of negligence, ignorant, and time constraint involved in awaiting an ophthalmologist for diagnosis or detection. Therefore, this study examined the diagnosis of long-sightedness using three algorithms which include Neural Network, Decision Tree, and Back Propagation, and this led to the development of an Expert System. Backpropagation and Decision tree algorithms were employed to train the Neural Network. A decision tree was implemented using a knowledge extraction rule to classify and categorise the disease based on the patient's symptoms. The C# programming language was used for the system implementation, and MySQL was used for the database. The outcome of the developed system explained how the sickness was identified so eradicating the impenetrability in the Neural network only, and finally, the plan was tested after development.