About the Course:

This comprehensive course provides an in-depth understanding of Data Analytics by exploring statistical modelling, regression, machine learning, and deep learning.

Through supervised, unsupervised and deep learning techniques, learners will gain insight into how to analyse complex datasets and unlock deeper insights and predictive analysis. The course is designed to empower learners to build data-driven systems that predict, recommend, and assist in a broad range of industries and applications.

By the end of the course, learners will have a robust understanding of data analytics, advanced machine learning, and deep learning concepts, thereby enhancing their analytical skills and professional development.

 

Entry Requirements:

Prior to enrolling in this course, learners are expected to have a working knowledge of the Python programming language, including variables and data types, operators, control flow, functions and using packages. Successful completion of the Professional Academy Certificate in Python Programming fulfils this requirement.

In addition to this, learners are also required to have working knowledge of data importing, cleaning, manipulation and visualisation. Successful completion of the Professional Academy Certificate in Data Analytics: Visualisation fulfils this requirement.

 

Resource Requirements:

● Operating System: Windows 10 or higher / Mac OS/ Linux
● RAM: 8 GB or higher for better multitasking and analysis of large size data
● Processor: 2.6 GHz or greater for faster processing (i5 or higher)
● Video Card: Integrated Graphic card or a better GPU for interactive visualisation practice
● A Google account is required to complete additional course exercises through Google Colab
● All other software (e.g. VS Code, Git, Jupyter, Anaconda) optional

 

Learning Outcomes:

Upon completion of this course, learners will be able to:

● Demonstrate a strong understanding of machine learning concepts, including supervised and unsupervised learning, ensemble methods, and neural networks.
● Implement machine learning techniques using Python and the Scikit-Learn library, including data preprocessing, model selection, evaluation, and hyperparameter tuning.
● Create advanced visualisations of data using tools like Bokeh, Matplotlib, and Seaborn to effectively communicate insights to stakeholders.
● Apply the CRISP-DM methodology to real-world datasets, including problem formulation, data preparation, model building, evaluation, and deployment.
● Effectively communicate results and insights to stakeholders through presentations, reports, and visualisations.

 

Course Structure Options:

● Part-time (evening): one evening per week; 6:30pm to 7:30pm*

*40 hours over 8 weeks. 4 hours self-study + 1 hour live expert training per week

– Scheduled classes: 8 hours
– Additional online instruction: 32 hours
– Self-study: 20 hours
– Assignment: 15 hours

Each week, our UCD Professional Academy lecturer will lead a 1-hour live online session, setting assignments for the week, answering your questions and giving you the guidance you need to progress in your career goals.

The other 4 hours (approximately) per week of learning are self-directed and can fit around your own schedule. The UCD Professional Academy has partnered with DataCamp to create bespoke learning tracks that give you the hands-on experience you need to master the skills of data analysis. DataCamp’s platform provides a state-of-the-art virtual environment, which allows you to simulate data analysis functions and receive instant feedback on your progress, without worrying about the technical limitations of your own equipment or purchasing expensive software. This approach also allows the lecturer to monitor your progress and any difficulties you encounter, and tailor the weekly check-in sessions appropriately.

 

Course Breakdown:

The following topics will be covered over a period of 8 weeks:

Module	Overview	Tools/Area covered
1. Introduction to Machine Learning	Machine learning definition, types of algorithms, workflow overview, and Scikit-Learn library introduction.	●       Definition and applications of machine learning ●       Types of machine learning algorithms ●       Overview of the machine learning workflow ●       Introduction to Scikit-Learn library
2. Advanced Statistical Learning	Data visualisation, hypothesis testing, correlation and regression analysis.	●          Interactive data visualisation with Bokeh ●          Statistical hypothesis testing ●          A/B testing ●          Correlation and regression analysis
3. Supervised Learning	Introduction to Supervised learning and types of Supervised learning. Model Selection.	●       Understanding supervised learning ●       Linear regression ●       Classification algorithms: logistic regression and KNN ●       Model selection and evaluation

4. Unsupervised Learning	Introduction to Unsupervised learning and types of Unsupervised learning. Model Selection.	●       Understanding unsupervised learning ●       Clustering algorithms: K-means and hierarchical clustering ●       Dimensionality reduction: PCA and t-SNE
5. Model Evaluation and Hyperparameter Tuning	Model Selection, Evaluation and Optimization.	●       Cross-validation and holdout method ●       Bias-Variance tradeoff ●       Hyperparameter tuning with grid search and randomized search. ●       Feature selection and engineering
6. Ensemble Methods	How to build and evaluate ensemble models, which can help improve the accuracy and robustness of their machine learning models.	●          Combining models: Voting, Bagging, and Boosting ●          Random Forests and Gradient Boosting Machines ●          Stacking and blending
7. Neural Networks	Artificial neural networks (ANNs) and their components, including activation functions and loss functions, as well as how to train them using backpropagation. Specific types of ANNs.	●       Introduction to Artificial Neural Networks (ANNs) ●       Activation functions and loss functions ●       Backpropagation and training ANNs ●       Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs)
8. Preparing the Final Project	Consolidate your understanding of the course materials and demonstrate your ability to apply machine learning techniques to solve a practical problem.	●          Applying the CRISP-DM methodology to a real-world dataset of your choice. ●          Implementing machine learning techniques learned throughout the course. ●          Presenting results and insights

 

Course Assessment:

Assessed Component	Weighting	Deadline
Final Project	100%	2 weeks after course completion

 

This course is assessed through the completion of a project, with supporting documentation, which is the application of machine learning models to a real-world dataset. The project requires the use of a dataset that is relevant to the student’s interests or work, provided it does not contain personal information.

This project should demonstrate the learner’s ability to apply the concepts and techniques learned in the course to a real-world problem, and to communicate the results and insights effectively.

In addition, DataCamp features a series of illustrative case studies which allow learners to develop and test their knowledge in a practical way, receiving instant feedback. There will also be additional exercises hosted in Google Colab. This constitutes the week-to-week ‘homework’ of the course but will not be graded.