MATHEMATICS FOR DATA SCIENCE
MAT4MDS
2019
Credit points: 15
Subject outline
Important mathematical ideas which underpin the theory and techniques of data science are introduced and consolidated in this subject. Matrices are used to store and work with quantitative information, and the methods of calculus are used to find extreme values and accumulation. The Gamma and Beta functions are introduced, as are eigenvalues, eigenvectors and the rank of a matrix. Emphasis is placed on the relevance of the mathematics to data science applications (such as least squares estimators and calculation of variance in data), and on the development of clear communication in explaining technical ideas. This is a foundational subject for the Master of Data Science.
School: School Engineering&Mathematical Sciences
Credit points: 15
Subject Co-ordinator: Katherine Seaton
Available to Study Abroad Students: No
Subject year level: Year Level 4 - UG/Hons/1st Yr PG
Exchange Students: Yes
Subject particulars
Subject rules
Prerequisites: Must be admitted in SMDS
Co-requisites: N/A
Incompatible subjects: MAT4NLA, MAT4CDE
Equivalent subjects: N/A
Special conditions: N/A
Graduate capabilities & intended learning outcomes
01. Perform mathematical calculations relevant to data science fluently and accurately.
- Activities:
- In the on-line modules and videos, the techniques will be demonstrated and explained. In the practice classes students will develop these skills.
02. Creatively apply mathematical techniques to unfamiliar problems.
- Activities:
- The practice classes are structured to move from consolidation of basic skills to the application of the techniques in a non-routine way.
03. Apply mathematical skills to analysis of data science literature.
- Activities:
- This will be demonstrated in the modules, and developed by engaging with the quizzes.
04. Present mathematical thinking in written form in a meaningful and succinct way using both words and mathematical notation.
- Activities:
- This is modelled in the modules, and specific instruction will be given in the practice classes. It is further developed through feedback on the assignments.
Melbourne, 2019, Semester 1, Blended
Overview
Online enrolment: Yes
Maximum enrolment size: N/A
Enrolment information:
Subject Instance Co-ordinator: Katherine Seaton
Class requirements
Unscheduled Online ClassWeek: 10 - 22
One 2.0 hours unscheduled online class per week on any day including weekend during the day from week 10 to week 22 and delivered via online.
"Readings and videos in LMS"
PracticalWeek: 10 - 22
One 2.0 hours practical per week on weekdays during the day from week 10 to week 22 and delivered via face-to-face.
Assessments
| Assessment element | Comments | % | ILO* |
|---|---|---|---|
| Four written assignments (750-words each, 3000-words total) | These assignments are problem-based and show consolidation of mathematical skills. | 35 | 01, 02, 03, 04 |
| Extended written task (1,500-words) | An analysis of an issue or paper in data science from the mathematical perspective, presented in an extended written form. | 15 | 03, 04 |
| Two hour exam (2,000-words equivalent) | 50 | 01, 02, 04 |