This data science project embarked on predicting obesity levels using machine learning techniques, leveraging a dataset encompassing attributes related to dietary habits and physical well-being across diverse populations. The investigation included the evaluation of various models, including Decision Trees, Random Forests, and Naive Bayes, in terms of their predictive accuracy and interpretability.

Remarkably, Decision Trees demonstrated comparable performance to Random Forests, while offering a more transparent decision-making process. Potential applications of this methodology include personalized healthcare and public health initiatives. It identifies factors that affect obesity, which could potentially improve the treatment process for different groups.

• Dataset: https://archive.ics.uci.edu/dataset/544/estimation+of+obesity+levels+based+on+eating+habits+and+physical+condition

• For more details, please refer to 'project.ipynb'