OCC Machine Learning Models Find Models that Have High Prediction Accuracy Ques

Problem 2 (25 points): In your own words, describe:

• the bias/variance tradeoff.
• two methods of selecting features.
• two approaches to compare/select models.
• a method for selecting the best K for your KNN algorithm
• a method for determining the best tree depth/ number of cases per parent node and number of cases per child node for your decision tree classifier.

Problem 3 (30 points): Sample of Data Science Interview questions

• What is machine learning?
• What is the difference between classification and regression?
• What are feature vectors?
• Explain the steps in building a decision tree.
• Explain Random Forest? (what are the steps in creating a random forest classifier?)
• What are two ways of reducing dimensionality.
• Explain ROC curve and what is AUC?
• What is bagging?
• What is K-means?
• What is overfitting?