C1000-059 Sample Questions Answers

they are a "blue sky thinker" who comes up with new approaches to use new data in innovative ways

they have a strong understanding of the enterprise's database architecture

they define data processes to meet compliance and regulatory obligations

the one who collects, processes, and performs statistical analysis on data

to randomize the trajectory of the neural network parameters during training

to smooth the gradient of the loss function in order to avoid getting trapped in small local minimas

to scale the gradient descent step in proportion to the gradient magnitude

to compute the gradient of the loss function with respect to the neural network parameters

increase the depth of the neural network

insert a dropout layer in the neural network architecture

increase the proportion of the train dataset by moving examples from the validation dataset to the train dataset

restart the training with a higher learning rate

replaces words with synonyms, e g. answer for reply

translates word by word

finds the root word

determines the category of a word, e.g nouns

It allows integration within business applications.

It makes it possible to create reports for management dynamically using specific parameters from executives.

It is critical for achieving high accuracy in training.

It is a necessary step in training and evaluating the performance of the models.

It is only possible on the cloud because they require a large amount of compute resources.

detecting objects in video streams

compressing large video files

aggregating sales revenue per state

creating a pivot table with monthly costs

division by random numbers

square

logarithm

subtract median

It is the probability of accepting a null hypothesis when the hypothesis is proven true.

It is the probability of rejecting a null hypothesis when the hypothesis is proven false.

It is the probability of accepting a null hypothesis when the hypothesis is proven false.

It is the probability of rejecting a null hypothesis when the hypothesis is proven true.

standard deviation = 0.5

expected value = 0

expected value = 0.5

expected value = 1

standard deviation = 1

apply dimensionality reduction to the features before training

impose an additional cost on the model for making classification mistakes on the minority class during training

lower the detection threshold of the minority class after training

oversample the minority class and/or undersample the majority class

after training, divide the model accuracy of each class by the proportion that they represent in the dataset

33 false positives and 81 false negatives

81 false positives and 73 false negatives

73 false positives and 81 false negatives

81 false positives and 33 false negatives

y = affine1(ReLU(affine2(x)))

y = max(affine1(x), affine2(x))

y = affine2(ReLU(affine1(x)))

y = affine2(max(affine1(x), 0))

y = ReLU(affine1(x), affine2(x))

paired t-test

chi squared test

z test

t-test

It is encoding for Latin script.

It is rarely used today.

It is encoding for Unicode characters.

It is equal to ASCII.

–9

17

12

–7