Predictive Analytics

Overview

This unit aims to develop students' conceptual and practical understanding of predictive analytics and machine learning for business application. The unit emphasises data-driven decision-making and the application of predictive modelling to address business challenges and opportunities across various industry sectors. Students will learn to leverage appropriate analytical techniques to analyse datasets, identify patterns, forecast trends, and build predictive models that inform strategic business decisions and optimise operations. Real-world case studies and industry applications will illustrate how predictive analytics drives innovation and delivers measurable impact across a range of business sectors.

Unit learning outcomes

Students who successfully complete this unit will be able to:

1. Evaluate the role of predictive analytics in business decision-making
2. Synthesise and apply predictive models to analyse business datasets, identify patterns, and generate actionable insights
3. Develop data-driven business solutions by applying appropriate machine learning techniques, considering the business rules and data requirements
4. Interpret and communicate complex analytical findings effectively using data visualisation for business stakeholders
5. Communicate effectively and professionally as either a leader or a member of a team in technology-driven projects

Teaching methods

Hawthorn

Assessment

Content

• The role, value, and industry applications of predictive analytics and AI in organisations.
• Data preparation and feature engineering: Data cleaning, transformation, handling missing values, and feature selection.
• Supervised learning for prediction and classification, e.g., Regression (linear, logistic) and classification (decision trees, random forests, SVMs).
• Unsupervised learning for pattern discovery, e.g., Clustering (K-Means, DBSCAN) and association rule mining.
• Time series forecasting and nusiness applications, e.g., ARIMA, exponential smoothing, Prophet, and business use cases.
• Model evaluation and performance optimisation, e.g., Performance metrics, overfitting, cross-validation, and hyperparameter tuning.
• Data Visualisation, effective reporting on analytics results, and communicating insights to stakeholders.
• Ethical AI and responsible use of Predictive Analytics, e.g., bias, privacy, transparency, and regulatory considerations in AI models.