CS-422 - Homework 3 (5%)

Classification

Due by: March 26, 2013

Assignment Specifications

In this assignment you will test and implement various classification techniques. Sample data files are available at http://archive.ics.uci.edu/ml/datasets/. The grade for this assignment will be based on your implementation of the algorithms, the thoroughness of your evaluation of the algorithms, the results you obtain, and the clarity of your report. Make sure to explain the results you obtain and do not unnecessarily repeat similar results. The code you write should be modular and well documented.

1. Use the spam email dataset from the previous assignment and identify one additional dataset to work with.

2. Use R to test several classification algorithms on your two datasets. Test the following algorithms (using the e1071, nnet, and caret packages): naive Bayes classifier, logistic regression, neural network, and support vector machines.

3. Compare the results you obtain with the different methods. Make sure to test different parameters in the algorithms. Draw conclusions from your comparison.

4. Implement and test the perceptron and logistic regression algorithms. Compare the results you obtain with the two classifiers and draw conclusions.

5. Compare the logistic regression algorithm when implemented with gradient descent and stochastic gradient descent. Draw conclusions.

General comments

• You are advised (but not required) to use R for implementing the logistic regression algorithm.
• Write your code in a modular way using functions and make sure to document it.
• Do not include in the submission large datasets that were provided by us.
• The assignment contains testing of many algorithms. Try to be concise and thorough in the way you present your results and make sure not to include repetitive results. Results you present should have a purpose.
• Follow the electronic submission instructions of assignment 1.