Q1. Analyzing music signals is used in the detection of emotion in music. In this case, music can be classified into several categories at the same time since they are not usually disjointed i.e. it can make you feel both sad and angry. A sample dataset that contains sound clips were annotated by three male music experts into six emotional clusters. Only the songs that had all three experts unanimously agree on their label were kept, resulting in a total of 593 songs being selected for the dataset. Artificial Neural Network (ANN) classifier was used to test this dataset. The following confusion matrix is provided as a result of this model with ROC Area of 0.946 Use the Confusion Matrix above to answer the following questions: a) Determine the accuracy and the error rate of the classifier (3 Marks) Accuracy = b) Calculate the Sensitivity/Recall rate for the positive class (2 Marks) c) Critically analyze and explain the usefulness of ANN classifier in predicting emotions in music using the above calculated performance measures. (2 Marks) d) The same dataset was used in cluster analysis and the K-means algorithm grouped the following 6 types of emotions into 3 clusters as follows As a business analytics expert, discuss the concept of unsupervised learning and why you think the data was clustered as such. (3 Marks) Jump to...