1
Supervised learning uses
Correct Answer
labelled training data
☑ MCQ PRACTICE
Data Analytics Unit 4
Practice objective questions for quick revision and examination preparation. Try answering each question before revealing the answer.
Data Analytics - Unit-4
2
Unsupervised learning
Correct Answer
finds patterns in unlabelled data
3
Which of the following is an unsupervised learning technique?
Correct Answer
K-means clustering
4
Segmentation is the process of
Correct Answer
dividing a population into groups of similar items
5
A classification tree is used when the target variable is
Correct Answer
categorical
6
A regression tree is used when the target variable is
Correct Answer
continuous
7
Which measure is commonly used to choose splits in a decision tree?
Correct Answer
Gini index or entropy
8
Information gain is based on the reduction in
Correct Answer
entropy
9
Overfitting occurs when a model
Correct Answer
fits the noise in the training data and generalises poorly
10
Pruning a decision tree is done to
Correct Answer
reduce complexity and avoid overfitting
11
Stopping the growth of a tree early is called
Correct Answer
pre-pruning
12
A random forest is
Correct Answer
an ensemble of many decision trees
13
Bagging stands for
Correct Answer
bootstrap aggregating
14
Boosting builds models
Correct Answer
sequentially, each one correcting the errors of the previous
15
In a decision tree, the terminal node that gives the prediction is the
Correct Answer
leaf node
16
A time series is
Correct Answer
a sequence of observations recorded in time order
17
ARIMA stands for
Correct Answer
AutoRegressive Integrated Moving Average
18
In ARIMA(p, d, q), d represents
Correct Answer
the degree of differencing
19
In ARIMA(p, d, q), p represents
Correct Answer
the order of the autoregressive part
20
A stationary time series has
Correct Answer
a constant mean and variance over time
21
RMSE stands for
Correct Answer
Root Mean Squared Error
22
MAPE stands for
Correct Answer
Mean Absolute Percentage Error
23
STL decomposition splits a time series into
Correct Answer
trend, seasonal and remainder components
24
In STL, the letters stand for
Correct Answer
Seasonal and Trend decomposition using Loess
25
Extracting features such as peak height and average energy from a signal is done to
Correct Answer
analyse the signal and support prediction
Fill in the Blanks
26
Learning from labelled data is called __________ learning.
Correct Answer
supervised
27
Clustering is an example of __________ learning.
Correct Answer
unsupervised
28
A regression tree predicts a __________ target variable.
Correct Answer
continuous
29
Decision trees choose splits using the Gini index or __________.
Correct Answer
entropy
30
__________ occurs when a model fits the noise in the training data.
Correct Answer
Overfitting
31
Removing branches from a tree to reduce its complexity is called __________.
Correct Answer
pruning
32
A random forest is an ensemble of many decision __________.
Correct Answer
trees
33
Bagging stands for __________ aggregating.
Correct Answer
Bootstrap
34
ARIMA stands for AutoRegressive Integrated __________ Average.
Correct Answer
Moving
35
In ARIMA(p, d, q), d is the degree of __________.
Correct Answer
differencing
36
A __________ time series has a constant mean and variance over time.
Correct Answer
stationary
37
RMSE stands for Root Mean __________ Error.
Correct Answer
Squared
38
STL decomposes a time series into trend, seasonal and __________ components.
Correct Answer
remainder
39
MAPE stands for Mean Absolute __________ Error.
Correct Answer
Percentage
40
The terminal node of a decision tree that gives the prediction is called the __________ node.
Correct Answer
leaf