Your goal is to evaluate the performance of a support vector machine (SVM) classifier trained on the Breast Cancer dataset from the UC Irvine repository. The classifier aims to predict whether there will be cancer recurrence events for treated breast cancer patients. The dataset was collected by the University Medical Centre, Institute of Oncology, Ljubljana, Yugoslavia. All questions in this homework will be based on the following table of actual class labels and predictions by the SVM for 20 data points in the test dataset which was held out from the classifier during training
Table 1: True class labels and predictions for the test set of the Breast Cancer dataset.
1. (a) Calculate the accuracy of the classifier on the above test set, expressed as a percentage. Recall that percentage accuracy is calculated as
(b) Briefly explain what is a baseline model?
(c) Briefly explain why is it important to compare the performance of a classifier to baseline models?
(d) Describe one simple baseline method which we could compare the performance of our SVM classifier to.
(c) Identify one limitation of classification accuracy as a measure of performance.
Students succeed in their courses by connecting and communicating with an expert until they receive help on their questions
Consult our trusted tutors.