Data Science is a rapidly expanding field in industry and many leading universities in the USA and the UK are starting data science degrees and units. Monash FIT trialed a data science unit for the undergraduate business IT course, FIT3152, in 2013, which is continuing, mostly focusing on data analysis. This Data Analysis unit is a core elective in the full Master of Data Science to start in 2016, and a core unit in the Graduate Diploma in Data Science (Monash Online) starting 2015.

23/2/2015: Amended wording in following fields for administrative clarification: Reasons For Introduction; Teaching/Mode; Assessment/Summary; Workload/Contact Hours; Location Of Offering

27/2/2015: Amended Workload Requirements following discussion with Chair of GPC to clarify the workload required for the various teaching modes.

12/05/2016: Amended the assessments and provided the rationale for having 100% in-semester assessments for the online and on-campus versions of the unit.

19/5/2016: The unit name is updated to what was agreed last year.

05/01/2017: Admin - adding RfC - update to on-campus assessment in line with University policy. Effective from Semester 1, 2017.

02/02/2018: Admin - As advised by the MDataSci Course Director in consultation with the CE, adding ETC5252 to the prerequisites will streamline take-up of the unit by BusEco students. ETC5252 is not equivalent to FIT5197 so these students also require experience with programming in R.

24/9/2019: Admin - adding 10 minutes reading time to the overall exam duration as per University requirements.

30/9/2020: Unit name changed to Machine learning, effective semester 1, 2020. With the introduction of the Master of Artificial Intelligence and changes made to the Master of Data Science, it was agreed that the name of this unit would be changed and are already reflected in the 2020 courses.

10/09/2020: Adding alternative prerequisite enrolment rule for students studying FIT5201 as an elective in S6001 Master of Financial Mathematics (MTH5530 and MTH5540). Discussed with Chief Examiner, S6001 Course Director and Deputy Dean Education (FIT). Both MTH units meet mathematical/statistical requirements, however enrolment will be offered on a case-by-case basis to those with sufficient R programming skills.

06/07/2021 Amended in-semester assessments % accordingly and keep consistency between hand book and Moodle unit preview.

This unit will be a core elective unit for the Master of Data Science starting 2016. It is also a core unit the online Graduate Diploma of Data Science starting 2015.

There are no existing masters units offering the same material as this unit. Although, FIT5197, Applied Data Analysis, has similar material but covers the use of algorithms rather than the theory behind them.

On successful completion of this unit a student should be able to:

1. describe what statistical machine learning and its theoretical concepts are.
2. assess a typical machine learning model and algorithm.
3. develop, and apply major models and algorithms for statistical learning.
4. scale typical statistical learning algorithms to learn from big data.

020119

This unit introduces machine learning and the major kinds of statistical learning models and algorithms used in data analysis. Learning and the different kinds of learning will be covered and their usage will be discussed. The unit presents foundational concepts in machine learning and statistical learning theory, e.g. , bias-variance, model selection, and how model complexity interplays with model?s performance on unobserved data. A series of different models and algorithms will be presented and interpreted based on the foundational concepts: linear models for regression and classification (e.g. linear basis function models, logistic regression, Bayesian classifiers, generalised linear models), discriminative and generative models, k-means and latent variable models (e.g. Gaussian mixture model), expectation-maximisation, neural networks and deep learning, and principles in scaling typical supervised and unsupervised learning algorithms to big data using distributed computing.

Christopher Bishop (2006). Pattern Recognition and Machine Learning. Springer.

On-campus, Monash Online

On-campus: Examination (2 hours and 10 minutes) 50%; in-semester 50%. Assignment 1: 25% Assignment 2: 16% Quiz: 9% Examination (2 hours and 10 minutes): 50%

Monash Online units cannot have exams, so there is a 100% in-semester assessment.

Minimum total expected workload equals 144 hours per semester comprising:

1. Contact hours for on-campus students:
   • Two hours/week lectures.
   • Two hours/week laboratories.
2. Contact hours for Monash Online students:
   • Two hours/week online group sessions.
   • Online students generally do not attend lecture, tutorial and laboratory sessions, however should plan to spend equivalent time working through resources and participating in discussions.
3. Additional requirements:
   • A minimum of 8 hours per week of personal study (22 hours per week for Monash online students) for completing lab/tutorial activities, assignments, private study and revision, and for online students, participating in discussions.

FIT

FIT5197 or (MTH5530 and MTH5540) or (ETC5252 plus experience with programming in R)