Decision Tree Algorithm and Gini Calculation for Puppies Skin Disease Diagnosis

Abstract

To put it another way, an expert system is a computer system that can store and reproduce knowledge in order to solve a problem. It is also possible to employ expert systems to assist a specific field with problem-solving, and the expert system serves to keep the necessary parties working together. Software or computer programs that provide guidance and assistance in specialized subjects, such as science, engineering, mathematics, medicine, and education can be referred to as expert systems. The expert system can be used to perform diagnostics on pets. This article aims to discuss the implementation and study of an expert system that assists users in determining the ailment that pets, particularly dogs, are suffering from. The system can generate a symptom diagnostic of a skin illness using a decision-tree algorithm. Veterinarians provided data sets on a variety of disorders. There were 105 cases of 11 diseases and 11 symptoms in the dataset used in this study. The scikit-learn library was also used to train decision trees. Following the creation of a decision tree, a case study analysis was conducted using ten real-world examples provided by the veterinarian. Participants who own dogs received questionnaires in addition to those used for analysis. In the study, 25.8% of respondents said they strongly agreed with the expert system, and 32.3% said they agreed that the expert system helped explain the ailment their dog had. A further 80 percent of the instances provided by vets can be accurately determined by the system. When four independent tests were run, the mean system prediction was found to be 67.6 percent.