Unit rationale, description and aim
The ability to predict future events and trends is crucial to success across industries. Predictive analytics uses historical data, statistical modelling, data mining and machine learning techniques to forecast future trends and events. It examines current and past data to identify patterns and trends. In this unit students will explore advanced techniques related to time series data by examining different business problems in a number of industries. This is one of the most common uses of predictive analytics and an important capability for data scientists. Students will gain an understanding of how predictive analytics can be used, and what the benefits, limitations and challenges are, particularly when using time series data. The aim of this unit is to give students the skills to support organisations in a range of industries to plan operations based on trends and likely future outcomes.
Campus offering
No unit offerings are currently available for this unit.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.
Critically evaluate the use of predictive analytic...
Learning Outcome 01
Design reproducible predictive analytic solutions ...
Learning Outcome 02
Evaluate the selection, implementation and use of ...
Learning Outcome 03
Communicate predictive analytics solutions to stak...
Learning Outcome 04
Content
Topics will include:
· Predictive Modelling - introduction
· Overview of Time Series Forecasting
· Pre-processing Time-Series
· Machine Learning for Time Series
· Techniques for Forecasting
· Deep Learning models
· Ensemble models
· Anomaly Detection in time series data
· Ethical AI and Bias Mitigation
· Further Models for Time-Series
· Future Trends in Predictive Analytics
Assessment strategy and rationale
Assessments are designed to ensure students gain a sound foundation of the complexities of predictive analytics. Assessment 1 has been designed to ensure that students gain a solid understanding of the role, techniques, benefits and limitations of predictive analytics. Working with a real-world data set, assessment 2 focuses on the process of developing a predictive model to answer a business-related research problem. Assessment 3 provides students an opportunity of exploring challenges of working with time-series data, reflecting on the process of developing a predictive model, and critically reviewing the selection, implementation and success of the project. This series of assessments scaffolds students learning by progressively increasing the complexity of tasks and requiring progressive integration of unit learning outcomes.
To pass the unit, students must demonstrate achievement of every unit learning outcome and obtain a minimum mark of 50%
Overview of assessments
Type – Report Purpose – Compare and contrast pr...
Type – Report
Purpose – Compare and contrast predictive analytic techniques, discussing their benefits and limitations in the role of decision making in a business setting.
This is an individual assessment
20%
Type –Report and presentation Purpose – Work wi...
Type –Report and presentation
Purpose – Work with real-world data sets to develop and evaluate a data predictive model to address a real-world business problem.
This is a group assessment
50%
Type – Report Purpose – Reflect on the process ...
Type – Report
Purpose – Reflect on the process undertaken to develop a predictive model, reporting on the selection and implementation, explaining why some algorithms worked well and others not so well, the successes and challenges, and the outcomes.
This is an individual assessment
30%
Learning and teaching strategy and rationale
The teaching approach within this unit puts the student at the centre of their learning. This is achieved by using a blended learning approach that integrates asynchronous interactive online elements with face-to-face learning experiences. Access to fundamental knowledge is provided through online resources that enable students to build their understandings in a flexible manner. Students are given the opportunity to build upon this knowledge through social learning experiences conducted in face-to-face classes such as tutorials and workshops. These opportunities enable students to build more complex understandings through peer interactions and structured learning experiences. This blended learning approach allows students to develop problem solving skills which align to vocational practices in data science.