School: Science

This unit information may be updated and amended immediately prior to semester. To ensure you have the correct outline, please check it again at the beginning of semester.

Please note that given the circumstances of COVID-19, there may be some modifications to the assessment schedule promoted in Handbook for Semester 1 2020 Units. Students will be notified of all approved modifications by Unit Coordinators via email and Unit Blackboard sites. Where changes have been made, these are designed to ensure that you still meet the unit learning outcomes in the context of our adjusted teaching and learning arrangements.

  • Unit Title

    Machine Learning and Data Visualisation
  • Unit Code

    MAT3120
  • Year

    2020
  • Enrolment Period

    1
  • Version

    2
  • Credit Points

    15
  • Full Year Unit

    N
  • Mode of Delivery

    On Campus
    Online
  • Unit Coordinator

    Dr Johnny Su Hau LO

Description

This unit introduces students to the principles and practices of machine learning to uncovering patterns and trends in complex data sets and then to visualising these patterns in meaningful ways. Machine learning is a process by which computer models are not explicitly programmed but "learn from data". Students will use existing data to develop models used to predict various outcomes for new data. Data may be derived from DNA sequencing, meteorological observations, social media, drug discovery, travel industry and much more.

Prerequisite Rule

Students must have passed MAT1114 Introductory Statistics or ECF1151 Quantitative and Statistical Techniques for Business or equivalent unit.

Learning Outcomes

On completion of this unit students should be able to:

  1. Demonstrate and communicate an understanding of the fundamental principles of machine learning and data visualisation.
  2. Critically assess the strengths and weaknesses of a range of machine learning methodologies as used in a range of applications.
  3. Select the machine learning algorithm most appropriate for a given real-world application and implement that method using existing computational libraries.
  4. Train a range of machine learning algorithms using a variety of big data sets and interpret the output.
  5. Objectively use a range of modern visualisation methods appropriate for different types of data.

Unit Content

  1. Principles of unsupervised and supervised machine learning.
  2. Model selection and feature selection.
  3. Model optimisation: cost functions, search space and other methods.
  4. Model evaluation and visualisation.
  5. Current machine learning methods to analyse and visualise large and complex data sets.

Additional Learning Experience Information

Laboratories, lectures, self-directed study.

Assessment

GS1 GRADING SCHEMA 1 Used for standard coursework units

Students please note: The marks and grades received by students on assessments may be subject to further moderation. All marks and grades are to be considered provisional until endorsed by the relevant Board of Examiners.

Due to the professional competency skill development associated with this Unit, student attendance/participation within listed in-class activities and/or online activities including discussion boards is compulsory. Students failing to meet participation standards as outlined in the unit plan may be awarded an I Grade (Fail - incomplete). Students who are unable to meet this requirement for medical or other reasons must seek the approval of the unit coordinator.

ON CAMPUS
TypeDescriptionValue
Laboratory WorkLaboratory exercises40%
PresentationOral presentation20%
ReportReport on analysis of a real data set40%
ONLINE
TypeDescriptionValue
Laboratory WorkLaboratory exercises40%
PresentationOnline presentation20%
ReportReport on analysis of a real data set40%

Disability Standards for Education (Commonwealth 2005)

For the purposes of considering a request for Reasonable Adjustments under the Disability Standards for Education (Commonwealth 2005), inherent requirements for this subject are articulated in the Unit Description, Learning Outcomes and Assessment Requirements of this entry. The University is dedicated to provide support to those with special requirements. Further details on the support for students with disabilities or medical conditions can be found at the Access and Inclusion website.

Academic Misconduct

Edith Cowan University has firm rules governing academic misconduct and there are substantial penalties that can be applied to students who are found in breach of these rules. Academic misconduct includes, but is not limited to:

  • plagiarism;
  • unauthorised collaboration;
  • cheating in examinations;
  • theft of other students' work;

Additionally, any material submitted for assessment purposes must be work that has not been submitted previously, by any person, for any other unit at ECU or elsewhere.

The ECU rules and policies governing all academic activities, including misconduct, can be accessed through the ECU website.

MAT3120|2|1

School: Science

This unit information may be updated and amended immediately prior to semester. To ensure you have the correct outline, please check it again at the beginning of semester.

Please note that given the circumstances of COVID-19, there may be some modifications to the assessment schedule promoted in Handbook for this unit. All assessment changes will be published by 27 July 2020. All students are reminded to check handbook at the beginning of semester to ensure they have the correct outline.

  • Unit Title

    Machine Learning and Data Visualisation
  • Unit Code

    MAT3120
  • Year

    2020
  • Enrolment Period

    2
  • Version

    2
  • Credit Points

    15
  • Full Year Unit

    N
  • Mode of Delivery

    On Campus
    Online
  • Unit Coordinator

    Dr Johnny Su Hau LO

Description

This unit introduces students to the principles and practices of machine learning to uncovering patterns and trends in complex data sets and then to visualising these patterns in meaningful ways. Machine learning is a process by which computer models are not explicitly programmed but "learn from data". Students will use existing data to develop models used to predict various outcomes for new data. Data may be derived from DNA sequencing, meteorological observations, social media, drug discovery, travel industry and much more.

Prerequisite Rule

Students must have passed MAT1114 Introductory Statistics or ECF1151 Quantitative and Statistical Techniques for Business or equivalent unit.

Learning Outcomes

On completion of this unit students should be able to:

  1. Demonstrate and communicate an understanding of the fundamental principles of machine learning and data visualisation.
  2. Critically assess the strengths and weaknesses of a range of machine learning methodologies as used in a range of applications.
  3. Select the machine learning algorithm most appropriate for a given real-world application and implement that method using existing computational libraries.
  4. Train a range of machine learning algorithms using a variety of big data sets and interpret the output.
  5. Objectively use a range of modern visualisation methods appropriate for different types of data.

Unit Content

  1. Principles of unsupervised and supervised machine learning.
  2. Model selection and feature selection.
  3. Model optimisation: cost functions, search space and other methods.
  4. Model evaluation and visualisation.
  5. Current machine learning methods to analyse and visualise large and complex data sets.

Additional Learning Experience Information

Laboratories, lectures, self-directed study.

Assessment

GS1 GRADING SCHEMA 1 Used for standard coursework units

Students please note: The marks and grades received by students on assessments may be subject to further moderation. All marks and grades are to be considered provisional until endorsed by the relevant Board of Examiners.

Due to the professional competency skill development associated with this Unit, student attendance/participation within listed in-class activities and/or online activities including discussion boards is compulsory. Students failing to meet participation standards as outlined in the unit plan may be awarded an I Grade (Fail - incomplete). Students who are unable to meet this requirement for medical or other reasons must seek the approval of the unit coordinator.

ON CAMPUS
TypeDescriptionValue
Laboratory WorkLaboratory exercises40%
PresentationOral presentation20%
ReportReport on analysis of a real data set40%
ONLINE
TypeDescriptionValue
Laboratory WorkLaboratory exercises40%
PresentationOnline presentation20%
ReportReport on analysis of a real data set40%

Disability Standards for Education (Commonwealth 2005)

For the purposes of considering a request for Reasonable Adjustments under the Disability Standards for Education (Commonwealth 2005), inherent requirements for this subject are articulated in the Unit Description, Learning Outcomes and Assessment Requirements of this entry. The University is dedicated to provide support to those with special requirements. Further details on the support for students with disabilities or medical conditions can be found at the Access and Inclusion website.

Academic Misconduct

Edith Cowan University has firm rules governing academic misconduct and there are substantial penalties that can be applied to students who are found in breach of these rules. Academic misconduct includes, but is not limited to:

  • plagiarism;
  • unauthorised collaboration;
  • cheating in examinations;
  • theft of other students' work;

Additionally, any material submitted for assessment purposes must be work that has not been submitted previously, by any person, for any other unit at ECU or elsewhere.

The ECU rules and policies governing all academic activities, including misconduct, can be accessed through the ECU website.

MAT3120|2|2