This project aims to develop a machine learning (ML) model to estimate obesity levels based on eating habits and physical condition.
[1] This dataset include data for the estimation of obesity levels in individuals from the countries of Mexico, Peru and Colombia, based on their eating habits and physical condition. The data contains 17 attributes and 2111 records, the records are labeled with the class variable NObesity (Obesity Level), that allows classification of the data using the values of Insufficient Weight, Normal Weight, Overweight Level I, Overweight Level II, Obesity Type I, Obesity Type II and Obesity Type III.
Description of the main parts of the project.
First, an Exploratory Data Analysis (EDA) was performed to determine and know some characteristics of the dataset to give a preliminary analysis of the behavior (accuracy) of the model that was made and Principal Components Analysis (PCA) was used to show the relationships between the attributes with the distribution of obesity levels.
For the creation of this model, firstly was performed a feature selection using Principal Components Analysis (PCA) and Gain Information (GI) based on [2] in order to improve model's accuracy. This approach was decided because of not all features are relevant to make a good prediction and also to avoid a overfitting caused by redundant information.
- [1] Estimation of Obesity Levels Based On Eating Habits and Physical Condition, UCI Machine Learning Repository, DOI: https://doi.org/10.24432/C5H31Z, 2019.
- [2] E. Odhiambo Omuya, G. Onyango Okeyo, and M. Waema Kimwele, “Feature selection for classification using principal component analysis and information gain,” Expert Systems with Applications, vol. 174, p. 114 765, 2021, issn: 0957-4174. doi: https://doi.org/10.1016/j.eswa.2021.114765.