Problem: Load the abalone sample dataset from the UCI Machine Learning Repository (abalone.data) into R using a dataframe. Remove all observations in the Infant category, keeping the Male/Female classes. Using the caret package, use createDataPartition to perform an 80/20 test-train split (80% training and 20% testing). Fit a logistic regression using all feature variables via glm, and observe which predictors are relevant. Do the confidence intervals for the predictors contain 0 within the range? How does this relate to the null hypothesis? Use the confusionMatrix function in caret to observe testing results (use a 50% cutoff to tag Male/Female) - how does the accuracy compare to a random classifier ROC curve? Use the corrplot package to plot correlations between the predictors. How does this help explain the classifier performance?