Brief
Write a 2-page report detailing the implementation and evaluation of a method for learning a classifier on the data provided. You are not required to code the classifier from scratch. Feel free to use a machine learning toolbox such as scikit-learn in Python. (Alternatively, you could also use Weka (in Java), shogun (in C++), or stats (in Matlab).
Data for training and testing your classifier are provided here:resitdata.zip
You should train your classifier using the dataset and evaluate its performance in an appropriate way. Note that you will be assessed on your report and on the code you produce.
Your report should include the following sections:
- APPROACH (maximum mark: 10)
You should present a high-level description and explanation of the machine learning approach (e.g., logistic regression, multi-layer perceptron, support vector machine) you have adopted in this re-sit coursework. You should give details of the way that the approach works and note any underlying theory and important assumptions on which it depends.
- METHODOLOGY (maximum mark: 30)
You should explain how you have gone about training and testing your classifier, including data pre-processing. Some discussion of the nature of the training data and any issues that arise from that should be included here. You should explain what data was used for training and how. You should also explain how performance of the classifier was assessed.
- RESULTS and DISCUSSION (maximum mark: 30)
You should present the results of your evaluation. There are different ways that this can be done (e.g., in terms of confusion matrices and relevant measures of performance; using graphs to show changing performance for different training sets, if you choose to do that, etc.). The main thing is to present the results sensibly and clearly. Critically evaluate the work and discuss any limitations or problems. You should also take the opportunity to discuss any ways you can think of to improve the work you have done. If you think that there are ways of getting better performance, then explain how. If you feel that you could have done a better job of evaluation, then explain how. What lessons, if any, have been learnt? Were your goals achieved?
- CODE APPENDIX (maximum mark: 30)
The whole program, with comments.