Artificial Intelligence in Business Applications Final Examination, Spring-2026

1. a) Describe how Artificial Intelligence (AI) acts as a "smart assistant" in business transformation, specifically highlighting its role in improving "Customer Experience (CX)" with real-world examples like Amazon. b) Explain the difference between Supervised and Unsupervised Learning based on the presence of "labeled data". Provide one business use case for each (e.g., Fraud detection vs. Customer segmentation). 2. a) Apply the K-Nearest Neighbors (KNN) algorithm to find the distance between two customer data points in a 2D space: Point A (3, 5) and Point B (7, 8). • Calculate the Euclidean Distance. • Calculate the Manhattan Distance. b) Illustrate the step-by-step process of how a Decision Tree works, starting from the "Root Node" to the "Leaf Nodes". Use a "Loan Approval" scenario (features: Income, Credit Score) as your example. 3. a) Describe a Customer Churn Prediction using the Logistic Regression formula z = B0 + B1. Scenario: A bank uses "Number of Complaints" (X) to predict churn. • • Given: Constant (Bo) = -1.5, Coefficient (B1) = 0.6, and a customer has 4 complaints. Task: Calculate the value of z and apply the Sigmoid function P(y=1) to find the probability of churn (Use e0.9 =2.46). b) Based on your calculation in (a), if the Decision Rule is P>= 0.5, will this customer leave the bank? Discuss how Logistic Regression differs from Linear Regression in terms of the "type of outcome" it predicts. 4. a) Solve the following transaction dataset and calculate the Support for the item-set {Bread, Milk}. • T1: Bread, Milk, Butter • T2: Bread, Butter • T3: Milk, Eggs • T4: Bread, Milk • T5: Bread, Milk, Eggs (Total Transactions = 5) b) Construct the Confidence for the Association Rule: {Milk} {Bread} using the dataset above. c) If the Minimum Confidence Threshold is 75%, compute whether the rule {Milk} -> {Bread} is a "Strong Rule". Suggest how a grocery store could use this rule for "Store Layout Planning".