Data Science Objective Set 1 – Cloud Education Courses

Which of the following is not considered a cause of confusion about the precise meaning of the data science buzzwords?

The constant evolution of the data science industry and in turn the meaning of the data science buzzwords

Ans:-The speed with which new data science terms are appearing

Which of the following is related to the meaning of the term analytics?

Analytics is about separating a dataset into easy-to-digest chunks and studying them individually and examine how they relate to other parts

Ans:-Analytics is the application of logical and computational reasoning to the component parts obtained in an analysis

Which of the terms relates to the field of business analytics only?

Creating dashboards

Reporting with visuals

Ans:- Qualitative analytics

Which of the following is not considered a data analytics activity?

Ans:-Business case studies

Preliminary data reporting

Optimization of drilling operations

Which of the following is considered data science?

Business case studies

Qualitative analytics

Digital signal processing

Sales forecasting

Given that all activities can be done with ML and all can be done without ML, choose the best answer. Which of the following is considered Data science but not Machine learning?

Creating real-time dashboards

Ans:-Sales forecasting

Fraud prevention

Which of the following is not an example of where Machine Learning is being applied today?

Ans:-Symbolic reasoning

Client retention

Image recognition

From a data scientist’s perspective, the solution of every task begins:

by suggesting a few hypothetical and theoretical solutions to your boss

by gathering your team and deciding on what approach to follow to solve the task

Ans:-with a proper dataset

According to our infographic, which of the following is not considered data science?

Ans:-Big data

Business intelligence

Traditional data science methods

Which of the following is related to the pre-processing of a traditional data set?

Class labelling

Data cleansing

Dealing with missing values

Ans:-All of the above

Which of the following do you encounter when working with big data?

Text data

Integer

Digital image data

Ans:-All of the above

The process of representing observations as numbers is called:

Collecting observations

Ans:-Quantification

A measure that has a business meaning attached is called:

an observation

a quantification

Ans:-a metric

A KPI (Key Performance Indicator) can be best defined as:

the accumulation of observations to show some information

Ans:-a metric that is tightly aligned with your business objectives

a quantification that has a business meaning attached

an observation that can potentially be related to the business goals of a company

The job of a business intelligence analyst always involves the creation of:

reports

dashboards

KPIs

Ans:-All of the above

Which of the following columns from our infographic contain activities that are said to belong to the field of ‘predictive analytics’ and do not aim at explaining past behaviour?

Traditional data

Big data

Business intelligence

Ans:-Traditional methods

In business and statistics, which is the general term that refers to using a model for quantifying causal relationships?

Ans:-regression analysis

factor analysis

cluster analysis

time-series analysis

Which technique can be implemented if you want to reduce the dimensionality of a certain statistical problem?

Ans:-factor analysis

cluster analysis

time-series analysis

all of the above

Which technique is associated with plotting values against time, shown always on the horizontal line?

regression analysis

Ans:-time-series analysis

factor analysis

cluster analysis

When the data is divided into a few groups, you should apply:

factor analysis

Ans:-cluster analysis

time-series analysis

Which of the following statements is true?

The core of machine learning is creating an algorithm, which a computer then uses to find a model that fits the data as best as possible

In machine learning, one does not give the machine instructions on how to find a model. Rather, one provides it with algorithms which give the machine directions on how to learn on its own

A machine learning algorithm is like a trial-and-error process, but the special thing about it, is that each consecutive trial is at least as good as the previous one

Ans:-All of the above

Which line represents the four ingredients of any machine learning algorithm?

Model, data, reward system, objective function

Ans:-Data, model, objective function, optimization algorithm

Model, labelled data, unlabelled data, optimization algorithm

Choose the best answer.

In which type of machine learning is one always working with unlabelled data?

Supervised learning

Ans:-Unsupervised learning

Reinforcement learning

In reinforcement learning, a reward system is being used to improve the machine learning model at hand. The idea of using this reward system is to:

to minimize the error of the model

to minimize the objective function

Ans:-to maximize the objective function

to improve the optimization algorithm

Which of the following is an example where big data techniques are being applied?

Basic customer data

Ans:-Social media

What is NOT a type of machine learning?

supervised

unsupervised

Ans:-reinforced

Which of the following is a typical real-life example where BI techniques are being applied?

Historical stock price data

Ans:-Inventory management

In geometrical terms, a scalar can be represented as:

a line

a square

Ans:-a point

Which of the following is a typical real-life example where machine learning techniques are being applied?

Ans:-Client retention

Basic customer data

Which of the following is a typical real-life example where traditional data techniques are being applied?

Ans:-Basic customer data

Inventory management

Which of the following is a TYPICAL real-life example where traditional data science techniques are being applied?

Social media

Financial trading data

Inventory management

Ans:-Sales forecasting

You have financial data for 100 countries. You feed them to the algorithm and ask it to classify them in as many groups as it sees fit. It starts with 100 groups as each country represents a separate group. You decide to tell it to spit out 5 major groups, i.e. cluster them in 5 clusters. This is an instance of:

supervised learning

Ans:-unsupervised learning

Which software tool is frequently used when working with traditional data or when doing a BI analysis?

Ans:-Excel

Hadoop

Knowing which programming language is a huge advantage if you are supposed to be working with big data and/or machine learning?

SQL

Ans:-Java

Data which can be classified using a linear model is called:

clusterable

regressable

linear

Ans:-linearly separable

Econometric time-series analysis is the domain of which software tool?

Excel

Ans:-E-views

Why do we express probabilities numerically?

To determine whether they are likely or unlikely.

Ans:-To compute which event is relatively more likely.

How do we graphically express two mutually exclusive sets?

Their circles are tangent to one another.

Ans:-Their circles do not touch.

What are the intersection and union of two mutually exclusive sets?

The intersection is their sum and the union is the empty set.

The intersection is the smaller set and the union is the larger set.

The union is the empty set and their sum is the intersection.

Ans:-The union is their sum and the intersection is the empty set.

If the probability of an event remains unaffected by another event, the two are….

Dependent.

Ans:-Independent.

What do we call the probability we use to distinguish dependent from independent events?

The dependent probability.

The independent probability.

Ans:The conditional probability.

What can you conclude about events A and B, given that P(A) = P(A|B)?

Ans:The two are independent.

The two are dependent.

What is the difference between P(A|B) and P(B|A)?

The former suggests the two events are dependent, while the latter suggests they are independent.

The former suggests event A is more likely than event B, while the latter suggests B is the likelier of the two.

Ans:One indicates the probability of getting A, given B has occurred, while the other indicates the likelihood of getting B, given A has occurred.

What is the value of P(A|B), knowing P(B|A) = 0.6, P(A) = 0.4 and P(B) = 0.3?

0.24

0.2

Ans:0.8

Which of these characteristics of relational databases are not typical of data warehouses?

Ability to store and query data

Quick retrieval of individual records

Use of structured data

Transaction processing

Answer:

Quick retrieval of individual records.

Use of structured data.

Transaction processing

Question:

Which of these is NOT one of the ACID properties of transactions?

Atomicity

Coordinated

Durability

Answer:- Coordinated

Question:

Match the following use cases to the type of processing system which is best suited for the task.

OLAP

OLTP

A:Money transfer from one account to another
B:Generate monthly sales reports
C:Spot trends in TV viewership over the week
D:Update a student’s test score in a final exam

Answer:- OLAP (B,C)

OLTP(A,D)

Question:

Which of these is NOT a data warehouse?

Amazon Redshift

Hive

Jenkins

Teradata

Answer:- Jenkins

Question:

Which of these is NOT a feature of Amazon Redshift?

Ability to back up data using snapshots

Interfaces to query and analyze data

Fully managed data warehouse on the cloud

Highly optimized transaction processing

Answer:- Highly optimized transaction processing

Question:

Which of the following are characteristics of distributed systems?

Fault tolerance

Software to coordinate tasks

Multiple machines working together

Difficulty in scaling

ans:-

Fault tolerance

Software to coordinate tasks

Multiple machines working together

Question:

What does an AWS Policy represent?

A collection of permissions on AWS resources

A licensing agreement between the user and Amazon Inc.

A collection of users who can create resources on AWS

A list of suggested use cases for each AWS service

ANS- A collection of permissions on AWS resources

Question:

When provisioning an AWS user account that will be used to access AWS programmatically, what is generated in order to authenticate users?

An AWS Secret Access Key

An AWS Access Key ID

A Kubernetes secret password

An @aws.com email address

Ans-

An AWS Secret Access Key

An AWS Access Key ID

Question: Which statements about the Amazon Redshift Query Editor is TRUE?

It can be used to load and query data, but not create tables

It can be used to query system tables

It can be used to create tables and to load and query data

It can be used to query existing data but not load data into tables

answer:-

It can be used to query system tables

It can be used to create tables and to load and query data

Question:

What AWS CLI command will list all the Redshift clusters which a user has access to?

aws redshift describe-clusters

aws redshift list-clusters

aws redshift view-all-clusters

aws redshift list-clusters –all

ans:- aws redshift describe-clusters

Question:

Which of the following is not true of Redshift clusters?

Users can define the VPC in which they will be provisioned

The cluster can be scaled

It supports the encryption of data on the cluster

The minimum cluster size is four nodes

ans:- The minimum cluster size is four nodes

Question:

Which of these features and metrics are offered by Amazon Redshift?

Cluster snapshots

Optimized transaction processing

Saved queries

Resource usage metrics

ans:-

Cluster snapshots

Saved queries

Resource usage metrics

Question:

Which of these details need to be supplied when provisioning a Redshift cluster using the Quick launch feature?

Cluster identifier

Cluster credentials

Node type

Frequency of automated snapshots

Question:

How is a SMOTE oversampler initialized from a ModelFrame object named mf?

mf.over_sampling.SMOTE()

mf.imbalance.SMOTE()

mf.imbalance.over_sampling.SMOTE()

mf.SMOTE()

Ans:- mf.imbalance.over_sampling.SMOTE()

Question:

Which statements are true concerning disabling automated snapshots and excluding tables from snapshots?

To exclude a table from a snapshot, add the NO BACKUP clause to the table definition

To disable an automated snapshot, set the retention period to 1 days

To disable an automated snapshot, change the retention period to 0 days

To exclude a table from a snapshot, use the BACKUP NO clause following the table definition

Ans:-

To disable an automated snapshot, change the retention period to 0 days

To exclude a table from a snapshot, use the BACKUP NO clause following the table definition

Question

Given the following values in the body_style column of a Pandas dataframe, what would they look like after label encoding?

hatchback, sedan, suv
(1, 0, 0), (0,1,0), (0,0,1)
 1, 2, 3
 (0, 1,1), (1,0,1), (1,1,0)
 0, 1, 2
Ans:-0, 1, 2


Question
 Which of these statements about the application of the RandomOversampler are correct?

Samples from the minority class are oversampled until the dataset is balanced

Samples from the majority class are dropped until the dataset is balanced

Samples for the minority class are generated to balance the data

The resultant balanced dataset contains duplicates

Ans:- Samples from the minority class are oversampled until the dataset is balanced

The resultant balanced dataset contains duplicates

Question

Match the following sampling techniques to the type of sampling they perform.

Oversampling ABCD

Undersampling ABCD

A:Near Miss
B:SMOTE
C:Tomek Links
D:Neighborhood Cleaning Rule

Answer :- Oversampling B

Undersampling ACD

Question

How do you obtain the explained variance ratios for the principal components in your data once you have fitted a PCA object to it?

pca.explained_variance_ratios

pca.get_variance_ratios()

pca.variance_ratios()

pca.explained_variance

ANS:- pca.explained_variance_ratios

Question

How do you initialize a principal component analyzer to extract five principal components from a ModelFrame named mf_data?

decomposition.PCA(data = mf_data, n_components = 5)

mf_data.decomposition.PCA(n_components = 5)

dimensionality_reduction.PCA(data = mf_data, n_components = 5)

mf_data.dimensionality_reduction.PCA(n_components = 5)

ANS:- mf_data.decomposition.PCA(n_components = 5)

Question

What property of a ModelFrame contains the name of the target column?

target_column

target

target_header

target_name

Ans:- target_name

Question

Given you have your test data in a ModelFrame called mf_test, and have a trained estimator called trained_model, how would you get the predictions from the model on your test data?

mf_test.predict(trained_model)

trained_model.predict(mf_test)

trained_model.make_predictions(mf_test)

trained_model.get_predictions(mf_test)

ANs:-mf_test.predict(trained_model)

Question

Which of these are characteristics of the EasyEnsembleClassifier and the BalancedRandomForestClassifier?

They use oversampling to balance the data

They use undersampling to balance the data

They use point-based sampling to balance the data

They use an ensemble of learners where each individual learner is weak

The number of learners can be configured

Ans:-They use undersampling to balance the data

They use an ensemble of learners where each individual learner is weak

The number of learners can be configured

Question:

Recall the essential types of cloud migration.

Re-platforming

Shift to SaaS

Shift to PaaS

Retooling

Ans:-

Re-platforming

Shift to SaaS

Question:

Identify the essential benefits of implementing big data analytics in the cloud.

Scalability and flexibility

Improved analytical outcomes

Security and privacy

Improved analysis

Ans:-

Security and privacy

Improved analysis

Question:

Which are essential characteristics of Kubernetes?

Kubernetes servers can run within a Docker container

Kubernetes is a powerful cloud container management and orchestration tools

Kubernetes servers can run within a Docker virtual machine

Kubernetes is a powerful cloud application management and orchestration tools

Ans:-

Kubernetes servers can run within a Docker container

Kubernetes is a powerful cloud container management and orchestration tools

Question: Select the IP addresses that are used in Kubernetes.

Process EndpointIP

InternalIP

ExternalIP

DataIP

Ans:-

InternalIP

ExternalIP

Question: Name some of the prominent tools provided by GCP that are used to ingest data.

Kubernetes Engine

Cloud Pub/Sub

BigQuery

Cloud Firestore

Ans:-

Kubernetes Engine

Cloud Pub/Sub

Question: Choose the big data management services provided by AWS.

RedShift

DataScale

ENR

EMR

Ans:-

RedShift

EMR

Question: Which of the following storages can we use to back up big data in AWS?

Amazon Glacier

Amazon RedShift

Amazon DataScale

Amazon S3

Ans:-

Amazon Glacier

Amazon S3

Question: Identify which of the following services and implementations can help facilitate disaster recovery planning in the cloud?

S3 configuration

Data migration

Backup and Restore planning

Multi-region deployments

Ans:-

Backup and Restore planning

Multi-region deployments

Question:

Identify some of the critical perspectives of the Cloud Adoption Framework that can help facilitate the implementation or adoption of cloud computing.

Service perspective

People perspective

Application perspective

Business perspective

Ans:-

People perspective

Business perspective

Question: What are the some of the critical layers involved in a typical blockchain-based cloud framework?

Cloud

Defining resources and parameters that are required for deployment

Cloud automation

Blockchain management

Ans:-

Cloud automation

Blockchain management

Question: What are the prominent blockchain implementations that we can use to manage distributed ledgers?

Datachain

Livecoin

Tron

Ethereum

Ans:-

Tron

Ethereum

Question: To which data lifecycle phase does importing data apply to?

Tutorial Contents