Machine Learning –Assignment 2 – 45% of Module

Based on the sample jupyter notebook files that you have been using in class you are required
to complete the following exercises. Using any datasets of your choice (but not the same ones
used in the examples) you should create your own jupyter notebook files for each of the
following Machine Learning techniques:

x a) Linear Regression Aim for originality and multiple datasets

lots of commentary of data, code and results
lots of explaination
x b) Polynomial Regression equal marks for each
lots of content online and in labs
c) Naive Bayes Classifier Kaggle for datasets

d) KFold Cross Validation Aim for interesting examples

over explain it
e) Decision Trees
can get good or bad results, thats okay but just explain why or what it
f) Support Vector Machine shows

marks if data unusual or a bit different

x g) K-Means Clustering
give source of data ie kaggle
h) K-Nearest Neighbour 9 data and 9 books

i) Principal Component Analysis CPU add blocks of lines to explain whats happening. not essays on
essays
explaination where expl req'd
so will be needed at start to outline what to do. then steps and
results.
Instructions why results may be right or wrongs and how could be improved

The jupyter notebook files should document what youadd comments and text cells
are doing and explain the results. The
purpose of this exercise is to show that you can understand, edit and run pre-existing code
samples applying ML techniques. Each file is expected to run and should explain the
technique. Each file (a) – (i) carries equal marks. Marks will be awarded for originality and for
clean, readable code with good explanations.
The code should be original and created/edited by the student. If the student is using code
from the practical sessions it should be explained thoroughly to demonstrate that the student
understands the code. You may be asked to explain your code if there are limited comments
and there are problems with originality. Results should also be explained thoroughly to
demonstrate understanding.
This assignment must be submitted by Wednesday 24th March 2021 by midnight. To submit,
a single zip file with your name on it should be emailed to eoghan.furey@lyit.ie (if the file is
too big then a link to cloud storage may be used).
Failure to submit by the deadline without a valid medical certificate will result in in a loss of
30% of the marks for this assignment. It is unfair on your class mates to expect that you can
have more time when they didn’t! Treat it like your job and get it in on time! NOTE: The
samples must be different from those supplied to you in blackboard.
email a zip folder with notebooks and datasets in sub folders
rather don't use sikit learn datasets have a go at each one and explain why did/didn't work
use kaggle or explain results

don't need fancy dataset. just try be a little unique

each part is equal, so start and focus on easier ones to get going