The Bayesian vs frequentist approaches: Implications for machine learning – Part One