Business Intelligence, Analytics, Data Science,
and AI, 5th Edition by Ramesh Sharda, Dursun
Delen, Efraim Turban
Chapter 7 Deep Learning and Cognitive Computing
1) In the opening vignette, banks had begun to use deep learning to provide more accurate
prediction of the customer churn.
Answer: FALSE
Diff: 2 Page Ref: 399
2) Deep learning is a separate branch of computational sciences from AI.
Answer: FALSE
Diff: 1 Page Ref: 403
3) Structured, human-mediated machine-learning approaches have been working fine for rather
abstract and formal tasks.
Answer: TRUE
Diff: 2 Page Ref: 403
4) The initial idea for deep learning began fairly recently, right around 2013.
Answer: FALSE
Diff: 2 Page Ref: 404
5) Neurons are processing units that perform a set of predefined mathematical operations on the
numerical values coming from the input variables or from the other neuron outputs to create and
push out its own outputs.
Answer: TRUE
Diff: 2 Page Ref: 406
6) In artificial neurons weight terms are adjustable but biased terms are fixed.
Answer: FALSE
Diff: 2 Page Ref: 406
7) The task undertaken by a neural network does not affect the architecture of the neural
network; in other words, architectures are problem independent.
Answer: FALSE
Diff: 2 Page Ref: 411-412
8) In neural network training, the output of the transfer function is fed into the summation
function.
Answer: FALSE
Diff: 3 Page Ref: 412
1
Copyright © 2024 Pearson Education Ltd.
,9) Neural network training is usually done by defining a performance function.
Answer: TRUE
Diff: 1 Page Ref: 415
2
Copyright © 2024 Pearson Education Ltd.
, 10) Deep neural networks are referred to as "deep" because they are primarily used to evaluate
substantially deeper philosophical issues.
Answer: FALSE
Diff: 1 Page Ref: 425
11) The growth in the use of deep neural networks has been spurred by advancements in GPU
hardware technology.
Answer: TRUE
Diff: 2 Page Ref: 425
12) Feedforward neural network models allow both forward and backward connections between
neurons.
Answer: FALSE
Diff: 2 Page Ref: 426
13) Neural networks are called "black boxes" due to the lack of ability to explain their reasoning.
Answer: TRUE
Diff: 2 Page Ref: 420
14) CNNs are deep learning architectures that are designed to process only the image data sets.
Answer: FALSE
Diff: 2 Page Ref: 430
15) ANN are capable of solving complex problems and provide detailed explanations of their
capabilities.
Answer: FALSE
Diff: 2 Page Ref: 419-420
16) In CNN type deep learning models, the convolution layers are often followed by a pooling
layer.
Answer: TRUE
Diff: 3 Page Ref: 431
17) Deep MLP and convolutional networks are specialized for processing a sequential grid of
values.
Answer: FALSE
Diff: 3 Page Ref: 442
18) RNN basically models a dynamic system with sequential inputs.
Answer: TRUE
Diff: 2 Page Ref: 442
19) Compute unified device architecture, which enables software developers to use GPUs, was
originally designed by Google.
Answer: FALSE
Diff: 2 Page Ref: 450
3
Copyright © 2024 Pearson Education Ltd.
and AI, 5th Edition by Ramesh Sharda, Dursun
Delen, Efraim Turban
Chapter 7 Deep Learning and Cognitive Computing
1) In the opening vignette, banks had begun to use deep learning to provide more accurate
prediction of the customer churn.
Answer: FALSE
Diff: 2 Page Ref: 399
2) Deep learning is a separate branch of computational sciences from AI.
Answer: FALSE
Diff: 1 Page Ref: 403
3) Structured, human-mediated machine-learning approaches have been working fine for rather
abstract and formal tasks.
Answer: TRUE
Diff: 2 Page Ref: 403
4) The initial idea for deep learning began fairly recently, right around 2013.
Answer: FALSE
Diff: 2 Page Ref: 404
5) Neurons are processing units that perform a set of predefined mathematical operations on the
numerical values coming from the input variables or from the other neuron outputs to create and
push out its own outputs.
Answer: TRUE
Diff: 2 Page Ref: 406
6) In artificial neurons weight terms are adjustable but biased terms are fixed.
Answer: FALSE
Diff: 2 Page Ref: 406
7) The task undertaken by a neural network does not affect the architecture of the neural
network; in other words, architectures are problem independent.
Answer: FALSE
Diff: 2 Page Ref: 411-412
8) In neural network training, the output of the transfer function is fed into the summation
function.
Answer: FALSE
Diff: 3 Page Ref: 412
1
Copyright © 2024 Pearson Education Ltd.
,9) Neural network training is usually done by defining a performance function.
Answer: TRUE
Diff: 1 Page Ref: 415
2
Copyright © 2024 Pearson Education Ltd.
, 10) Deep neural networks are referred to as "deep" because they are primarily used to evaluate
substantially deeper philosophical issues.
Answer: FALSE
Diff: 1 Page Ref: 425
11) The growth in the use of deep neural networks has been spurred by advancements in GPU
hardware technology.
Answer: TRUE
Diff: 2 Page Ref: 425
12) Feedforward neural network models allow both forward and backward connections between
neurons.
Answer: FALSE
Diff: 2 Page Ref: 426
13) Neural networks are called "black boxes" due to the lack of ability to explain their reasoning.
Answer: TRUE
Diff: 2 Page Ref: 420
14) CNNs are deep learning architectures that are designed to process only the image data sets.
Answer: FALSE
Diff: 2 Page Ref: 430
15) ANN are capable of solving complex problems and provide detailed explanations of their
capabilities.
Answer: FALSE
Diff: 2 Page Ref: 419-420
16) In CNN type deep learning models, the convolution layers are often followed by a pooling
layer.
Answer: TRUE
Diff: 3 Page Ref: 431
17) Deep MLP and convolutional networks are specialized for processing a sequential grid of
values.
Answer: FALSE
Diff: 3 Page Ref: 442
18) RNN basically models a dynamic system with sequential inputs.
Answer: TRUE
Diff: 2 Page Ref: 442
19) Compute unified device architecture, which enables software developers to use GPUs, was
originally designed by Google.
Answer: FALSE
Diff: 2 Page Ref: 450
3
Copyright © 2024 Pearson Education Ltd.
