Question

a) Let’s suppose you work in a Bank sector in which one of the main problem is to stop the fake transactions. As we know, organizations are increasingly turning towards advanced analytics and machine learning algorithms to identify consumption patterns. Now as you are the senior developer, you have to design a classification model which will indicate a fake transaction. However, one of the biggest stumbling blocks is the humongous data and its distribution. Fraudulent transactions are significantly lower than normal healthy transactions. As a senior developer, identify those problems which will be faced during this process, if any? And what will be the steps or methods, you will use to resolve it. Explain in detail. b) Suppose that you have a data set that contains multi collinearity and high dimensionality. Explain which technique you will consider to overcome this problem. c) What is Stratification? Discuss its role in K-fold cross validation techniques. d) Discuss the role of weak learner in Ensemble-Algorithms. Explain, why Boosting algorithms are considered stronger as compared to Bagging algorithms when data is noise free (without outlier).

Answer 1

Answer:

a / b c / go to golden book