In this experiment, we will use a custom R module to compute Baseline Metrics for a Binary Classifier to evaluate our classification model.
When we build a classification model, often we have to prove that the model we built is significantly better than random guessing. How do we know if our machine learning model performs better than a classifier built by assigning labels or classes arbitrarily (through random guess, weighted guess etc.)? In this experiment, we will use a custom R module to compute Baseline Metrics for a Binary Classifier against which we can evaluate our classification model. Baseline metrics can be used as benchmarks with which to compare the performance of machine learning models and provide a justification for the results obtained from the model even when the numbers suggest otherwise. For more details please read http://blog.revolutionanalytics.com/2016/03/classification-models.html <img src="https://raw.githubusercontent.com/shaheeng/ShaheenGauher/master/Images/BaselineMetricExpScreenshot.png" width="300" height="400" /> Fig. Screenshot of the experiment <img src="https://raw.githubusercontent.com/shaheeng/ShaheenGauher/master/Images/BaselineMetricmoduleinput.png" width="300" height="400" /> Fig. Input parameters for the module <img src="https://raw.githubusercontent.com/shaheeng/ShaheenGauher/master/Images/BaselineMetricmoduleoutput.png" width="200" height="400" /> Fig. Module output containing Baseline Metrics for a Binary Classifier