Wealth Class Dataset

Import the dataset city_persons.csv to your PyCharm environment

I ended up using Jupyter Notebook since pycharm was giving me some issues. To import the data into Jupyter Notebook I simply copied the data link and then downloaded a csv file to my desktop. I then used pandas to create a dataframe.

Initially set the target to the least wealthy class, 2 in this case, and set all other wealth class outcomes to 0 (3,4 & 5)

This step was simple. The only thing that had to be changed from the provided script was setting the target value equal to 1.

Train, validate and test your model

Training the model required the column names to be changed. This dataset was not too hard to work with because the variables were numeric. I used bucketized columns (every 10 years) for age and indicator columns for the rest of the variables.

Interpret and analyze your results. Did the model performance exhibit a particular trend?

The two images above show the results for setting all the wealth classes equal to the target. As you can see, wealth class 2 was very accurate, and wealth class 3 did well. Then the data got far less accurate for wealth classes 4 and 5.