concrete-compressive-strength

• This was part of a class project where different variables are used to predict the concrete compressive strength
• Exploratory Data analysis was performed to clean the data
• Data was normalized using Z-score approach
• Algorithms used: Lasso and Ridge Regression
• Models were developed using the two ML algorithms and for each model, different values were assumed for the penalty function
• The effect of different values of penalty parameter on feature selection was studied.
• 5-fold validation was used to chose the penalty parameter for the predictive model based on Lasso Regression.