 

Khoa học dữ liệu cho kinh doanh

(DAT722_241_11_L01)
Nhà của tôi / Các khoá học của tôi
/ 2024-2025 / Học kỳ I
/ Khoa Khoa học dữ liệu trong kinh doanh
/ Khoa học dữ liệu cho kinh doanh
(DAT722_241_11_L01)
/ Exercises and Tests
/ Test 5 - Linear Regression

Bắt đầu vào lúc Thứ Sáu, 22 Tháng Mười

Một 2024, 10:14 SA
State Finished
Kết thúc lúc Thứ Sáu, 22 Tháng Mười
Một 2024, 10:26 SA
Thời gian thực 12 phút 7 giây
hiện
Ðiểm 28,00/30,00
Ðiểm 9,33 out of 10,00 (93%)

Câu hỏi 1

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. Which of the following terms determines

the proportion of variance in the dependent
variable that can be explained by the
independent variable?

Select one:
a. Variance

b. R-Squared 

c. Mean

d. Standard deviation

Câu hỏi 2

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. Suppose that we have a dataset D and we

design a linear regression model of degree 3
polynomial and we found that the training and
testing error is “0” (perfectly fits). What will
happen when you fit a degree 2 polynomial in
linear regression?

Select one:
a. Cannot say

b. None of these

c. There are high chances that degree 2

polynomial will overfit the data

d. There are high chances that degree 2

polynomial will underfit the data 

Câu hỏi 3

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. According to the lecture, the house price

is determined as follows: price = Theta_0 +
Theta_1*size. Theta_0 and Theta_1 are called
...

Select one:
a. Dependent variable

b. Hypothesis

c. Model's parameters 

d. Independent variable

Câu hỏi 4

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. Which of the following is NOT a type of

regularization used in linear regression?

Select one:
a. Elastic Net

b. Decision Tree 

c. Lasso

d. Ridge

Câu hỏi 5

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. The square root of the average square

differences between the data and the mean
value is called ...

Select one:
a. Standard deviation 

b. Mean

c. R-Squared

d. Variance

Câu hỏi 6

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. In linear regression, which of the following

assumptions is NOT required?

Select one:
a. Multicollinearity 

b. Homoscedasticity

c. Linearity

d. Normality of residuals

Câu hỏi 7

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. Suppose that the linear regression model

is underfitting the data. In such a situation,
which of the following regularization
algorithms would you prefer?

Select one:
a. L2

b. Any

c. None of the mentioned methods 

d. L1

Câu hỏi 8

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. According to the lecture, the house price

is determined as follows: price = Theta_0 +
Theta_1*size. This equation is called ...

Select one:
a. Independent variable

b. Hypothesis 

c. Model's parameters

d. Dependent variable

Câu hỏi 9

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. What would be the root mean square

training error for the data set that has 3
instances: (1,1), (2,2), and (6,6) if a Linear
Regression model of the form (Y = A0+A1X) is
applied?

Select one:
a. 0 

b. 2

c. 1

d. square(13)

Câu hỏi 10

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. According to the lecture, assume that we

just use a single feature to predict house
price (e.g., price = Theta_0 + Theta_1*size),
which of the following statements is correct?

Select one:
a. If Theta_0 = 0, then the model may output
zero if the house size is greater than 0

b. If Theta_1 = 0, then the model always

outputs a constant value regardless of house
size 

c. If Theta_1 is positive, then the price and

the size are in the opposite direction

d. If Theta_1 is negative, then the price and

the size are in the same direction

Câu hỏi 11

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. According to the lecture, the house price

is determined as follows: price = Theta_0 +
Theta_1*size. The term 'size' is called ...

Select one:
a. Model's parameters

b. Hypothesis

c. Independent variable 

d. Dependent variable

Câu hỏi 12

Sai

Ðạt điểm 0,00 trên 1,00

Ðặt cờ

L3. In Lasso regression, the penalty term

added to the loss function is based on which
norm?

Select one:
a. L2 norm

b. L-infinity norm

c. L0 norm 

d. L1 norm

Câu hỏi 13

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. Suppose that we have a dataset D and we

design a linear regression model of degree 3
polynomial and we found that the training and
testing error is “0” (perfectly fits). What will
happen when we fit a degree 4 polynomial in
linear regression?

Select one:
a. Cannot say

b. None of these

c. There are high chances that degree 4

polynomial will overfit the data 

d. There are high chances that degree 4

polynomial will underfit the data

Câu hỏi 14

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. The domain of Root Mean Square Error

(RMSE)?

Select one:
a. [0, infinity], the larger the better

b. [0,1], the closer to zero, the better

c. [0, infinity], the smaller the better 

d. [0,1], the closer to one, the better

Câu hỏi 15

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. The validation set, which is a part of the

original, preprocessed data set, is primarily
used to ...

Select one:
a. Initially fit the model

b. Provide an unbiased evaluation of an

initial model and tune the model's hyper-
parameters 

c. Provide an unbiased evaluation of a final

model before launch

d. Keep balance between data in different

subsets

Câu hỏi 16

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. The average of the square differences

between the data and the mean value is
called ...

Select one:
a. Mean

b. R-Squared

c. Standard deviation

d. Variance 

Câu hỏi 17

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. Suppose that the relationship between

independent variables x1, x2 and dependent
variable y is as follows: y = x0 - 0.95x1 +
0.96x2. Which of the following is true?

Select one:
a. Relation between x2 and y is neutral

b. Correlation can’t judge the relationship

c. Relation between x2 and y is strong 

d. Relation between x2 and y is weak

Câu hỏi 18

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. According to the lecture, if we use two

features to predict the house price (e.g., price
= Theta_0 + Theta_1*x_1 + Theta_2*x_2),
then the graph of the cost function is ...

Select one:
a. a surface in a 2-dimensional space

b. a surface in a 3-dimensional space

c. a surface in a 4-dimensional space 

d. a surface in a 1-dimensional space

Câu hỏi 19

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. Which method is typically used to

estimate the coefficients in linear regression?

Select one:
a. Maximum Likelihood Estimation

b. Newton's Method

c. Gradient Descent

d. Least Squares Method 

Câu hỏi 20

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. The domain of R-Squared?

Select one:
a. [0, infinity], the smaller the better

b. [0,1], the closer to one, the better 

c. [0, infinity], the larger the better

d. [0,1], the closer to zero, the better

Câu hỏi 21

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. Which of the following techniques can be

used to address multicollinearity?

Select one:
a. Adding more variables to the model

b. Increasing the sample size

c. Removing highly correlated predictors 

d. Using polynomial regression

Câu hỏi 22

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. In Ridge regression, the penalty term

added to the loss function is based on which
norm?

Select one:
a. L-infinity norm

b. L2 norm 

c. L0 norm

d. L1 norm

Câu hỏi 23

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. Linear Regression is a ... machine learning

algorithm.

Select one:
a. both supervised and unsupervised

b. unsupervised

c. supervised 

d. reinforcement

Câu hỏi 24

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. The process of selecting, manipulating,

and transforming raw data into features that
can be used efficiently by supervised learning
algorithms is usually referred to as ...

Select one:
a. Model training

b. Parameter tuning

c. Model testing

d. Feature engineering 

Câu hỏi 25

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. What is the purpose of using

regularization techniques in linear regression?

Select one:
a. To reduce overfitting by adding a penalty
to large coefficients 

b. To ensure the model is unbiased

c. To increase the model complexity

d. To improve the model's performance on

the training data

Câu hỏi 26

Sai

Ðạt điểm 0,00 trên 1,00

Ðặt cờ

L3. Suppose that the relationship between

independent variables x1, x2 and dependent
variable y is as follows: y = x0 - 0.95x1 +
0.96x2. Which of the following is true?

Select one:
a. Relation between x1 and y is weak

b. Correlation can’t judge the relationship 

c. Relation between x1 and y is neutral

d. Relation between x1 and y is strong

Câu hỏi 27

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. The train set, which is a part of the

original, preprocessed data set, is primarily
used to ...

Select one:
a. Provide an unbiased evaluation of an
initial model and tune the model's hyper-
parameters

b. Initially fit the model 

c. Provide an unbiased evaluation of a final

model before launch

d. Keep balance between data in different

subsets

Câu hỏi 28

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. In linear regression, what does the

intercept term represent?

Select one:
a. The predicted value of the dependent
variable when all independent variables are
zero 

b. The slope of the line

c. The average value of the dependent

variable

d. The error term

Câu hỏi 29

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. Choose the correct statement about

Residuals.

Select one:
a. Higher is better

b. Lower is better 

c. None of the mentioned statements

d. Lower or higher depends on the situation

Câu hỏi 30

Ðúng

Ðạt điểm 1,00 trên 1,00

Ðặt cờ

L3. Suppose that the linear regression model

is underfitting the data. In such a situation,
which of the following options would you
consider: 1. add more variables; 2. start
introducing polynomial degree variables; 3.
remove some variables?

Select one:
a. 1 and 3