Description
Overview
Collaboration: Do your work and report individually. You can collaborate on the right tools to
use and setting up your programming environment.
Hand in: One report per person, via the LEARN dropbox. Also submit the code / scripts needed
to reproduce your work. Report as a PDF or a python notebook.
General Objective: To practice performing classification on a small dataset.
Specific Objectives: • Establish your software stack to carry out data analysis homeworks,
assignments and the project for the rest of the course.
• Load a simple dataset and perform some basic data preprocessing, take out only the
attributes which are continuous or have distances defined.
• Divide data into train and test portions, justify your split decision.
• Perform classification and plot the results along some meaningful dimensions, analyse
the results.
Tools: You can use libraries available in Python or R available to you. You need to mention which
libraries you are using, any blogs or papers you used to figure out how to set carry out your
calculations.
Data sets
For this homework you will use the Bank Marketing dataset:
• https://archive.ics.uci.edu/ml/datasets/Bank+Marketing
• Download from Data Folder link, read data set description.
Tasks
In the class we talked about how to use k-Nearest Neighbours and SVM algorithms to classify
supervised data.
• Load the dataset and store it as a DataFrame.
• Formulate the problem as kNN and SVM (linear) problem and run using standard libraries
in your language or choice.
1
• Create a few plots of your model on the test data two of the data dimensions indicating
the predicted elements of each class using different colours or shapes. You may need to try
plotting various pairs of dimensions to see which provide some interesting result. Be sure to
label your axis and legend.
• Provide a short explanation of the result you have shown and what it means.
• For fun : do the same for SVM with RBF kernal and compare the results for the same
dimensions.
2
