Upgraded Machine Learning Techniques to Forecast Obesity Hazards

In recent years, investigators have prolonged their research on predicting obesity risks using Machine Learning (ML) techniques. Obesity is an intricate condition with multiple contributing factors, such as genetics, behaviour, environment, and social factors. Predicting obesity risk through ML can aid recognise individuals who may be at risk also aids in early identification to prevent or manage the condition. There is a need for these predictive models which is a combination of various factors like age, medical history, and lifestyle and these are the probability of increasing obesity among individuals. The model aims to accurately predict the obesity risks as early as possible. Detecting obesity at an initial stage is a stimulating procedure. This paper aims to use ML models such as XGBoost, CatBoost and LightGBM to predict obesity at a primary stage. To predict the obesity, Nobeyesdad dataset from Kaggle is used in this study. The obesity prediction model classifies the data into multiple categories like normal, insufficient weight, overweight and obesity individuals. The results of the enhanced obesity prediction model generates ideal results by classifying the datasets which aids the specialists to make decisions easily.