Building a predictive model to predict song popularity for 5 popularity levels in the dataset based on features.
The song popularity column takes the following values: {1,2,3,4,5}, in which the higher the number the more popular the song is. 5 means very popular and 1 means less popular.
Technologies
pandas
numpy
matplotlib
sklearn
Algorithms
Decision Tree Classification
Pearson Correlation
Project Objectives
Use feature scaling and train a decision tree model.
Using the test set, predict using the decision tree and compute the confusion matrix and
the accuracy of classification.
Report the findings from the confusion matrix and accuracy