Results

Accuracy is a metric that measures how often a machine learning model correctly predicts class labels[15].

Advantages

Gives good idea of how well a model performs at predicting on unseen data samples[16].

Generally easy to explain and communicate[17]

Considerations

The main limitation of the accuracy machine learning metric is that it can be biased when there is a clear imbalance in the data. If that is the case, the accuracy calculation mostly lean towards the majority class as the model that keeps predicting the majority class would have great accuracy[18.

As the main goal of our project is to produce a machine learning model that can classify credit scores as accurately as possible, we decided to use the accuracy machine learning metric to evaluate our finalized model. The confusion matrices below show how accurately the model classified credit scores of the 10% of the data used for testing the model after being trained with the 90% of the data used for training the model during each fold of the setup KFold Cross-Validationused to evaluate our finalized model. The way to calculate the accuracy of each of those matrices by hand is by adding the numbers in the left diagonal of the matrix over all the numbers in the matrix added together.

Now, when putting all these accuracy results together, we were able to conclude that our finalized model performed relatively well in classifying credit scores with a median accuracy of approximately 84.07% and a mean accuracy accuracy of approximately 84.11%.

Page updated

Google Sites

Report abuse