EE3700 Introduction to Machine Learning 1
HW4
1. For the Car Evaluation Data Set from UCI website (http://archive.ics.uci.edu/ml/
datasets/Car+Evaluation)
1.1. Combine class value “unacc” and “acc” as one group, “good” and “vgood” as the
other group, develop a model to identify the two groups.
1.2. With the original four class values, develop a model to do multi-class classification of
the four groups. (All features used)
1.3. Use feature selection or extraction to develop a model for four group classification.
Please also prepare some analysis for the model developed in 1.2 and 1.3
(performance, computation complexity required, …).
In the above questions, use all features to train the model. Please provide enough information
for the model development and testing. For example:
1. Preprocessing of dataset (categorical data handling, dataset partitioning, feature scaling,
…)
2. Why you choose the algorithm to develop the model? How do you select the parameters
3. Performance of the model you developed.
4. …
