03 09 18

Pagina 1

Politecnico di Milano School of Industrial and Information Engineering Data Mining and Text Mining September 3, 2018 FAMILY NAME FIRST NAME CODICE PERSONA/ID GENERAL INSTRUCTIONS • Answers must be clearly written inside the answer box designated for each. All the answers must be adequately motivated. • Pencils are not allowed. The exam consists of 6 sheets of paper. It must be returned with all the 6 sheets. No any other sheet can be added. No sheet can be removed. • This is a closed-book/closed-notes exam. • Only non-programmable calculators are allowed. • Notes/books/mobile phones are not allowed. • If you are caught using forbidden material, the exam will immediately end and an RP grade will be recorded; then, your Data Mining exam will consist of an oral examination from then on. SCORING • A problem left unsolved will amount to zero points. • A completely wrong solution will amount to -3 points STUDENTS WHO DID THE COURSE PROJECT HAVE 1:40h TO SOLVE PROBLEMS 1, 2, 3, AND 4 ALL THE OTHER STUDENTS HAVE 2:20h TO SOLVE ALL THE SIX PROBLEMS COURSE PROJECT SCORE FINAL TIME GRADES 1 2 3 4 5 6 Problem 1 (6 points). (1) Write the pseudo code for building a regression model using Gradient Boosting given the data X, the target variable y, and the number of predictors k. The code returns an array of k models “learner” and an array of k values of alpha. Note that, the value of alpha for the k-th predictor should be computed as 1/k. (2) Write the pseudo code for computing the prediction of a Gradient Boosting model given an array of predictors, an array of alpha values, and the input x. The code returns the value of the model in x. Notes a) Write one instruction per line. b) Write the algorithm using pseudo code. Python is not required but each line must be implementable by…