Artificial Intelligence in Business Applications – Spring 2025 – Midterm

Daffodil International University Department of Business Admiration Faculty of Business & Entrepreneurship Midterm Examination. Fall-2025 Course Code: 0613-324, Course Title: Artificial Intelligence in Business Applications Batch: 64 Teacher: TW Time: 1 Ilour and 30 Minutes Answer Questions Marks: 25 [The figures in the right margin indicate thefull marks and corresponding outcomes. All portions Ofeach question must be ansurred sequentially.] Let us consi&r a "here and are "i:h glasgs_ There is an AC (Air conditi.merj placed in orthe u hid System 'P' this is thx thc Current TN p If the is by air. If Set the S description Of em p 10 its agent in IS , Values for h: CLOI B 6 S z Demonstrate cniform cost search to find the solution pth frøm S to D. Interpret a search tree using A • alsotPithm to find thc from S •o page I of2 @CamScanner• Artificial (A Differentiate on principles. applieaöns. 131 121 4. You are designing an delivery robot logistics . Thc rotvt must navig•tc city deliver to cugomer locata•ns. During &sign "cussions. yow team deb*s u •s si»uld observabk and ot observable and stochastic. h ich Of City demo msuerby comparing it •ith a fully observable, deter"istic 121 environment uch •s a chess automated "Ming Page 2 of2 @CamScanner• Daffodil International University Department of Business Admiration Faculty of Business & Entrepreneurship Final Examination, Fall-2025 Course Code: 0613-324, Cqurse Title: Artificial Intelligence in Business Applications Batch: 64 Time: 2 Ilours Teacher: TW MFII Marks: 40 Answer ALL Questions [Thefigures in the right margin indicate the full marks and corresponding course outcomes. All portions Ofeach question must be answered sequentially.] A telecom company wants to reduce customer chum. They collected data on monthly usage, contract length, and monthly bill amount. They also classify each customer as "Likely to Churn" or "Not Likely to Churn." Using this business scenario, explain how the company can: 10 CLOI 2. • Use regression to predict monthly bill amount for new customers Use correlation to identify whether usage or contract length has a stronger relationship with churn behavior why regression and classification serve different business objectives Your answer must be business-focused with proper reasoning. b7nk uses an A1 model to classify loan applicants as ligh Risk or Low Risk. After testing, the confusion matrix is: Predicted Iligh Risk Predicted Low Risk Actual High Risk Actual Low Risk 25 30 225 10 Compute the following with full steps: l. Accuracy 2. Error Rate 3. True Positive (TPR) 4. me Negative Rate (TNR) Finally, explain in business terms what these metrics tell the bank about the reliability Of the loan approval model. page I Of 2 CL02 @CamScanner• 3. 4. Clothing retailer nantS to a nc•u• customer purchase a Premium or Regular product based On similarity past customers. masct: Customer c Incomc(kS) 30 SO 28 75 Scorc 40 60 30 70 New customer: Income 60k. Score 78 Using k = 3 (Euclidean distance): • Compute distance from each existing customer • Identify 3 nearest neighbors Product Regular Premium Premium Regular Premium 10 CL03 • Predict whether the new customer will choose Premium or Regular • Explain wily k'NN is useful for business customer segmentation All steps must be shown clearly. A fashion retail chain wants to increase sales using A1. Illey collect transaction daia from their Stores. Thc transactional dataset is: • T l: Sneakers} T4: (Jack:et, Sneakers) • T5: {T-Sbirt Jeans) Using this dataset: I. Generate frequent item-sets using Apriori (minimum support 40%). 2. Extract two Strong association rules (minimum confidence 6Wo). 3. Describe how deep learning creates value in fashion retail through: Demand forecasting Dynamic pricing Personalized style recommendations Trend prediction page 2 of2 10 CL04 @CamScanner•