Databricks Certified Professional Data Scientist Exam Practice Test Instant Access

Question 1

In which of the following scenario you should apply the Bay's Theorem

AThe sample space is partitioned into a set of mutually exclusive events {A1, A2, . .., An }.

BWithin the sample space, there exists an event B, for which P(B) > 0.

CThe analytical goal is to compute a conditional probability of the form: P(Ak | B ).

DIn all above cases

Answer : D

Question 2

You have used k-means clustering to classify behavior of 100, 000 customers for a retail store. You decide to use household income, age, gender and yearly purchase amount as measures. You have chosen to use 8 clusters and notice that 2 clusters only have 3 customers assigned. What should you do?

ADecrease the number of measures used

BIncrease the number of clusters

CDecrease the number of clusters

DIdentify additional measures to add to the analysis
kmeans uses an iterative algorithm that minimizes the sum of distances from each object to its cluster centroid, over all clusters. This algorithm moves objects between clusters until the sum cannot be decreased further. The result is a set of clusters that are as compact and well-separated as possible. You can control the details of the minimization using several optional input parameters to kmeans, including ones for the initial values of the cluster centroids, and for the maximum number of iterations.
Clustering is primarily an exploratory technique to discover hidden structures of the data: possibly as a prelude to more focused analysis or decision processes. Some specific applications of k-means are image processing^ medical and customer segmentation. Clustering is often used as a lead-in to classification. Once the clusters are identified,
labels can be applied to each cluster to classify each group based on its characteristics. Marketing and sales groups use k-means to better identify customers who have similar behaviors and spending patterns.

Answer : C

Question 3

You are creating a Classification process where input is the income, education and current debt of a customer, what could be the possible output of this process.

AProbability of the customer default on loan repayment

BPercentage of the customer loan repayment capability

CPercentage of the customer should be given loan or not

DThe output might be a risk class, such as 'good', 'acceptable', 'average', or 'unacceptable'.
Classification is the process of using several inputs to produce one or more outputs. For example the input might be the income, education and current debt of a customer The output might be a risk class, such as 'good', 'acceptable', 'average', or 'unacceptable'. Contrast this to regression where the output is a number not a class.

Answer : D

Question 4

What is one modeling or descriptive statistical function in MADlib that is typically not provided in a standard relational database?

AExpected value

BVariance

CLinear regression

DQuantiles
Linear regression models a linear relationship of a scalar dependent variable y to one or more explanatory independent variables x to build a model of coefficients.

Answer : C

Question 5

Regularization is a very important technique in machine learning to prevent over fitting. And Optimizing with a L1 regularization term is harder than with an L2 regularization term because

AThe penalty term is not differentiate

BThe second derivative is not constant

CThe objective function is not convex

DThe constraints are quadratic
Regularization is a very important technique in machine learning to prevent overfitting. Mathematically speaking, it adds a regularization term in order to prevent the coefficients to fit so perfectly to overfit. The difference between the L1 and L2 is just that L2 is the sum of the square of the weights, while L1 is just the sum of the weights.
Much of optimization theory has historically focused on convex loss functions because they're much easier to optimize than non-convex functions: a convex function over a bounded domain is guaranteed to have a minimum, and it's easy to find that minimum by following the gradient of the function at each point no matter where you start. For non-convex functions, on the other hand, where you start matters a great deal; if you start in a bad position and follow the gradient, you're likely to end up in a local minimum that is not necessarily equal to the global minimum.
You can think of convex functions as cereal bowls: anywhere you start in the cereal bowl, you're likely to roll down to the bottom. A non-convex function is more like a skate park: lots of ramps, dips, ups and downs. It's a lot harder to find the lowest point in a skate park than it is a cereal bowl.

Answer : A

Question 6

Scenario: Suppose that Bob can decide to go to work by one of three modes of transportation,

car, bus, or commuter train. Because of high traffic, if he decides to go by car. there is a 50% chance he will be late. If he goes by bus, which has special reserved lanes but is sometimes overcrowded, the probability of being late is only 20%. The commuter train is almost never late, with a probability of only 1 %, but is more expensive than the bus.

Suppose that Bob is late one day, and his boss wishes to estimate the probability that he drove to work that day by car. Since he does not know Which mode of transportation Bob usually uses, he gives a prior probability of 1 3 to each of the three possibilities. Which of the following method the boss will use to estimate of the probability that Bob drove to work?

ANaive Bayes

BLinear regression

CRandom decision forests

DNone of the above
Bayes' theorem (also known as Bayes' rule) is a useful tool for calculating conditional probabilities.

Answer : A

Question 7

You are working on a problem where you have to predict whether the claim is done valid or not. And you find that most of the claims which are having spelling errors as well as corrections in the manually filled claim forms compare to the honest claims. Which of the following technique is suitable to find out whether the claim is valid or not?

ANaive Bayes

BLogistic Regression

CRandom Decision Forests

DAny one of the above
In this problem you have been given high-dimensional independent variables like texts, corrections, test results etc. and you have to predict either valid or not valid (One of two). So all of the below technique can be applied to this problem.
Support vector machines Naive Bayes Logistic regression Random decision forests

Answer : D