Machine Learning with Python

TITLE Machine Learning with Python
MODULE CODE
CREDITS 10
NFQ LEVEL 9
MODULE DESCRIPTION This course deals with machine learning using Python. The student will learn where and how to apply Machine Learning algorithms. The course will cover examination of data and the range of tools available through Python for the analysis and visualisation of data. The students will deal with problems using supervised and unsupervised learning approaches and will learn to evaluate the models. Solutions will be found to make predictions for a range of datasets.
LEARNING OUTCOMES
1 To understand problems which are suitable for Machine Learning approaches
2 To store, manipulate, visualise and analysis data using the appropriate libraries and tools in Python.
3 Select the appropriate variable type and libraries for the problems under investigation and analysis of the data sets.
4 To apply statistical analysis to datasets
5 To explain, apply and evaluate a range of supervised and unsupervised learning approaches used to develop machine leaning solutions for datasets.
6 To be able to select the appropriate machine learning approach for a problem, and then design, implement, evaluate and document the solution.
INDICATIVE CONTENT
Python Fundamentals, Applications and Packages for Data Science Functions. Lists, Tuples, Sets, Dictionaries. Loops. List Comprehension. Error Handling File input and output. NumPy. Pandas Scikit-learn.
Intro to Machine Learning History, Evolution and Applications of Machine Learning. Understanding data and desired prediction. Machine Learning categories; Supervised, Unsupervised, Semi-Supervised, Reinforcement Learning.
Examination of Data with Statistics Understanding the raw data and attributes Determining statistical metrics of data. Data distribution analysis
Visualisation of Data Univariate plots. Multivariate plots.
Supervised Learning Regression and Classification. Linear Regression. Data Pre-processing techniques. Labels and Feature selection. Regressing Training and Testing. Regressing Forecasting and Predicting. Best fit line and R squared. Classification application using K Nearest Neighbours. Logistics regression. Support Vector Machines. Decision Trees.
Unsupervised Learning Types of cluster formation. K-means algorithm. Hierarchical Clustering.