A detailed report was made after developing 17 machine learning models using three distinct datasets for the tasks; unsupervised learning, regression and classification. The process involved data preprocessing including data engineering, feature selection and exploratory data analysis in order to answer specific research questions which were then compared with the existing literature. Principle component analysis was conducted along with Kmeans clustering, Dendogram, GMM and DBSCAN for the analysis of penguin species. Six regression models were developed to predict car price prediction and the most effective model was identified to be Random Forest Regressor. Six Classification models were developed to predict Airplane passenger satisfaction in which the most effective model was identified to be XG Boost.