Question 12 {5 points} A visa agency build a machine learning (ML) model to decide if a person is eligible to get a visa or not. The features of the ML model are; age. gender. nationality, country of residence. previous visasI # of times entering the country in the past 10 years, countries visited during the past 3 years. They will use 5000 labelled historical data records and their visa output to train this model. Which one of the following machine learning algorithms could fit best to build this model? 0 a} Classification algorithm {e.g., naive Bayes) O b} Clustering algorithm {e.g., K-means} O 1:} Regression algorithm {e_g_, Tree regression) Q d) All of the above Question 13 (5 points) A manufacturing company is interested in building a datavmining model of customer profitability based upon a series of independent variables {attributes}. [Customer profitability means the profitability of a customer is at a certain level or not.) The attributes include customer transaction history, demographics, and externally purchased credit-scoring information. There are currently 100,000 unique customers available for use in building this model. Which one of the following three techniques is suitable here: Clustering {e.g.. K-means}. Classication (e.g., Decision Tree or Naive Bayes} Regression {e.g.. Linear Regression or Regression Trees} 0 a} Clustering {e.g., K-means} O b} Regression (e.g., linear regression or regression trees) 0 c} All of the three techniques: classification, clustering and regression 0 d) Classification (e.g., Decision Tree or Naive Bayes}