Descriptive Analytics and Pattern Discovery (ANL306)

ANL306 Descriptive Analytics and Pattern Discovery equips students with practical skills in applying unsupervised learning techniques to uncover meaningful patterns and relationships in data. The course covers association rule mining, clustering, and other methods for developing descriptive analytics solutions to business and industry problems. Students will learn to construct, interpret, and evaluate unsupervised learning models using Python and translate the resulting patterns into actionable insights. By the end of the course, students will be able to apply appropriate pattern-discovery techniques across a range of real-world contexts.

• Introduction to Association
• Data Preparation for Association Analysis
• Association Rule Mining: Apriori
• Visualisation of Association Rules
• Introduction to Clustering
• Data Preparation for Clustering
• Proximity Measure
• Dimension Reduction
• Partitional Clustering
• Hierarchical Clustering
• Density-based spatial clustering of applications with noise (DBSCAN)
• Anomaly Detection – Local Outlier Factor

• Discuss various aspects of applying descriptive analytics and pattern discovery
• Appraise the application of an unsupervised learning method in a specific context
• Compare different techniques for pattern discovery using unsupervised learning methods
• Construct analytics models/results, including the use of AI-assisted approaches, as part of solutions to pattern discovery
• Evaluate the performance of descriptive analytics models
• Interpret the results of descriptive analytics models for addressing business problems
• Verify the validity, reliability and limitations of AI-assisted unsupervised learning models and outputs