Module Description
Introduction to Data Science introduces learners to the essential steps involved in understanding, preparing, and using data for machine learning. The module begins with descriptive statistics, showing how datasets can be summarized through data types, distributions, measures of central tendency, variability, missing values, outliers, and visualizations. It then moves into data preparation, including feature selection, categorical data transformation, cross-validation, and scaling. Finally, the module explains the process of building and evaluating machine learning models, including model comparison, hyperparameter optimization, and final performance evaluation using appropriate metrics for classification and regression problems.
Learning Outcomes
Participants will be able to:
- Explain the role of descriptive statistics in understanding the main characteristics of a dataset before applying machine learning.
- Distinguish between categorical and quantitative data and relate them to classification and regression problems.
- Apply exploratory data analysis techniques, including summary statistics and visualizations, to identify patterns, distributions, missing values, outliers, and relationships between variables.
- Describe key data preparation steps, including feature selection, categorical data transformation, cross-validation, and scaling.
- Explain how cross-validation helps evaluate model performance, reduce overfitting, and improve confidence in model generalization.
- Compare different machine learning models and understand how hyperparameter optimization can improve model performance.
- Evaluate final machine learning models using suitable metrics, such as precision, recall, ROC-AUC, accuracy, RMSE, MAE, and R-squared.
