ITEC613 - Introduction to Data Science

Year

Credit points

Campus offering

Prerequisites

ITEC610 Python Fundamentals for Data Science

Teaching organisation

3 hours per week for twelve weeks or equivalent.

Unit rationale, description and aim

Data science is an inter-disciplinary area that employs scientific methods, algorithms, tools and systems for extracting insights, knowledge and value from data. Machine learning, as a core part of data science and data analytics, and a subset of artificial intelligence, is the scientific study of algorithms and mathematical models that computer systems use to make decisions or predictions. Machine learning algorithms and models are widely used in human’s digital life such as email client, search engine, social media, virtual personal assistant, healthcare and recommendation system, although machine bias is an important ethical concern of which many people are unaware. Python is one of the most popular and globally adopted programming languages with comprehensive libraries and tools for putting data science and machine learning into practice in an efficient manner.

This unit will cover fundamental theories of data science and machine learning and their practical implementations and use while considering the issue of machine bias and how it may have an adverse impact on the common good. The aim of the unit is to provide students with fundamental data science and machine learning theories and techniques, together with popular tools to solve real-world problems such as in digital health.

Learning outcomes

To successfully complete this unit you will be able to demonstrate you have achieved the learning outcomes (LO) detailed in the below table.

Each outcome is informed by a number of graduate capabilities (GC) to ensure your work in this, and every unit, is part of a larger goal of graduating from ACU with the attributes of insight, empathy, imagination and impact.

Explore the graduate capabilities.

On successful completion of this unit, students should be able to:

LO1 - Demonstrate comprehensive knowledge of using data science libraries and tools for data processing and analysis (GA5, GA10)

LO2 - Appraise the use of fundamental data science and machine learning theories, key techniques and relevant tools for machine learning preparation (GA5, GA8)

LO3 - Develop an end-to-end data science and machine learning solution to real-world problems eg in digital health with appropriate choices of data science and machine learning techniques (GA4, GA5, GA7)

LO4 - Examine the issue of machine bias and how it may affect the common good (GA2, GA5, GA7)

Graduate attributes

GA2 - recognise their responsibility to the common good, the environment and society

GA4 - think critically and reflectively 

GA5 - demonstrate values, knowledge, skills and attitudes appropriate to the discipline and/or profession 

GA7 - work both autonomously and collaboratively

GA8 - locate, organise, analyse, synthesise and evaluate information 

GA10 - utilise information and communication and other relevant technologies effectively.

Content

Topics will include:

• Overview of data science and its implementation life cycle and tools
• Recap of data processing concepts including data quality and data operations such as cleaning, integration, reduction and transformation.
• Theory and practice of essential statistics in data science
• Machine learning (ML) introduction
• ML projects and basic linear algebra
• Basic matrix analysis and SVD, PCA
• Basic classification and evaluation with ROC curves
• Probability Theory and Naïve Bayesian Classifier
• Regression (linear, polynomial), overfitting and regularization, Bayesian regression
• Clustering: k-means and mixture of Gaussians
• Better evaluation with k-fold cross validation and finetune model with grid search
• Machine bias in the real world and its impact on the common good

Learning and teaching strategy and rationale

This unit will be delivered in a multimode over a twelve-week semester or equivalent study period. Students will have access to all primary learning materials online through LEO, along with formative and summative assessments, all of which will be available online, to provide a learning experience beyond the classroom. While there are no formal classroom lectures for this unit, students will be required to attend weekly two-hour workshop and fortnightly two-hour lab for the achievement of the unit learning outcomes. Workshops facilitate learning by theory comprehension and problem solving while lab sessions focus on hands on practices, which in combination is particularly effective for learning information technology skills.

Students should anticipate undertaking 150 hours of study for this unit, including class attendance, readings, online forum participation and assessment preparation.

Assessment strategy and rationale

A range of assessment procedures will be used to meet the unit learning outcomes and develop graduate attributes consistent with University assessment requirements. The first assessment consists of simple programming tasks to practice data processing, analysis practical and machine learning skills. The purpose is to assess students’ practical skills of using Python data science and machine learning libraries and tools for data processing and analysis. The second assessment is a more specific image data exploration and machine learning preparation task that covers fundamental knowledge of data science and machine learning. The purpose is to assess students’ understanding and skills in data preparation for machine learning preparation. The final assessment is a group project to do experiments with machine learning models and algorithms. The purpose is to assess students’ knowledge and skills of applying key machine learning algorithms to solve real-world problems e.g. in digital health with consideration of machine bias, continuing from the machine learning preparation task. There are fortnightly lab sessions associated with the assessments including assessable lab participation/engagement.

The assessments for this unit are designed to demonstrate the achievement of each learning outcome. To pass this unit, students are required to:

1. attempt all three assessment items 
2. obtain an overall mark of at least 50%

Overview of assessments

Representative texts and references

Aurélien Géron, 2019. Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd Edition, O'Reilly Media, Inc.

Christopher Bishop, 2006, Pattern Recognition and Machine Learning, Springer-Verlag New York.

EMC Education Services (Editor), 2015. Data Science and Big Data Analytics: Discovering, Analyzing, Visualizing and Presenting Data, Wiley.

Peter Bruce et al, 2020. Practical Statistics for Data Scientists, 2nd Edition O'Reilly Media, Inc.

Hadrien Jean, 2020. Essential Math for Data Science, O'Reilly Media, Inc.

Luca Massaron and John Paul Mueller, 2019. Python for Data Science, 2nd Edition, For Dummies.

Gilbert Strang, 2016. Introduction to Linear Algebra, fifth edition, http://math.mit.edu/~gs/linearalgebra/

Kavin P. Murphy, 2012, Machine Learning: A Probabilistic Perspective, MIT Press Academic

Vikas Kumar, 2018. Healthcare Analytics Made Simple: Techniques in healthcare computing using machine learning and Python, Packt Pulishing limited.

M. Ghassemi, T. Naumann, P. Schulam, A. L. Beam, I. Y. Chen, and R. Ranganath, 2020. A Review of Challenges and Opportunities in Machine Learning for Health, AMIA Joint Summits on Translational Science proceedings, vol. 2020, pp. 191-200.

C. Verdonk, F. Verdonk, and G. Dreyfus, 2020, How machine learning could be used in clinical practice during an epidemic, Critical Care, vol. 24, no. 1, p. 265.

Kubben et al (Eds), 2019. Fundamentals of Clinical Data Science, Springer – open access freely available from https://www.springer.com/gp/book/9783319997124