Business Analytics in the Age of Generative AI

Module one: The evolution of business analytics and what this means for you
● Understand the progression and evolution of data science from analytics to Machine Learning and more recently Generative AI 
● Recognise the communicative power of analytics and its role in decision-making
● Consider the evidence for Generative AI’s capabilities, strengths and limitations
● Explore different state of the art Generative AI models and understand the importance of prompt engineering 
● Address concerns related to data hallucinations and the importance of verifying insights
● Discuss data privacy issues with generative AI and ways to overcome related challenges
● Engage practically with ChatGPT, loading data, and conversing to derive actionable insight

Module two: Crafting data narratives through visualisation 
● Distinguish between visualisation as a discovery tool and as a tool to communicate insights
● Describe the principles of good data visualisation and some of the pitfalls
● Explore chat GPT and its “code interpreter” capability for efficient visualisations 
● Understand and apply storytelling principles using data
● Engage in hands-on practice of visualisation techniques with real-world datasets

Module three: Measuring the relationship  between variables 
● Present examples on how firms generate value using supervised learning
● Discuss three key applications for supervised learning: inference, prediction, classification
● Understand the model of linear regression and the ordinary least squares algorithm 
● Quantify the relationship between variables, the notions of statistical error, confidence intervals, and p-values 
● Apply the model of linear regression on large real world datasets 
● Use the framework of linear regression to conduct A/B testing 
● Appreciate why correlation does not imply causality 
● Recognise the importance of transparency and reproducibility in the hypothesis testing process, and master documentation workflow 

Module four: Learning from history to predict the future
● Explore relationships between multiple variables within linear regression framework
● Understand how to control for confounding variables 
● Engage in feature engineering to enhance predictive capabilities
● Become familiar with decision tree-based models and random forests
● Apply linear regression techniques to make predictions

Module five: Building predictive classification models
● Extend the model of linear regression to make predictions for categorical variables
● Understand the fundamentals of logistic regression and the maximum likelihood algorithm
● Understand risk scoring, its relevance, and how it’s used in classification
● Master the interpretation and application of the confusion matrix, ROC (Receiver Operating Characteristic) and AUC (Area Under the Curve) metrics
● Understand the challenges in prediction like overfitting and distinguish between training vs. testing datasets

Module six: Decoding data secrets: from pattern discovery to real-world application
● Understand and apply nearest neighbour algorithms for various use-cases
● Dive deep into clustering algorithms, focusing primarily on k-means and hierarchical clustering methodologies
● Develop an understanding of how datadriven recommendation engines operate
● Synthesise the skills and techniques learned throughout the course in a comprehensive capstone case
● Learn how two industry experts working in different industries are leveraging analytics to make an impact