Obesity disease risk prediction using machine learning

Obesity is a global issue that harms everyone’s health, regardless of age or gender. An excessive or abnormal amount of body fat is referred to as obesity. People who are obese are more likely to contract certain illnesses. Obese people are more likely to suffer from diabetes, thyroid, heart disease, liver cancer, and stroke, among other significant illnesses and ailments. The paper aims to gather raw data, perform data preprocessing on an obesity dataset to eliminate inconsistencies, and then transform the data into a format that can be more readily and efficiently handled by machine learning approaches. In the proposed paper, data preparation is carried out, including data integration, data reduction, and data cleaning. This work shows various visualizations using the obesity feature dataset. Support vector machines, random forests, and decision trees are used for classification. The random forest model has the best prediction accuracy (96%), followed by the decision tree and support vector machine models, according to the experiment analysis results.