代做AD699: Data Mining for Business Analytics Spring 2019 Quiz #3代做留学生Matlab编程

AD699: Data Mining for Business Analytics Spring 2019

30APR

Quiz #3

Version:  HOTEL

You have one hour to complete this quiz.  You may use a calculator, along with your book and/or notes, but may not use a smartphone or anything else with Internet connectivity.

For any multiple choice question, you are not being asked to choose the “best” from among four possibilities; instead,  there  are  three  wrong  answers,  and  one  right  answer.    Any  multiple  choice  question  must be answered with one completely clear answer choice.

For any free response questions, show your work.  Rounding is completely okay (and showing your work helps me to see what you did).

Do not pass a calculator or any other material to any student during the quiz.  If you do this, the quiz will end immediately for you and for the other student.

Free response questions that ask for multiple pieces of info will be scored in a binary fashion (1 or 0 points). There are three versions of this quiz, but all contain the same content.

The data science team at Target recently studied a series of consumer transactions at one of the company’s retail locations in Massachusetts.   Here is a list of the items purchased in a series of 15 recent transactions.  One row represents one consumer’s purchases at the register in one visit.

1.    Oatmeal, Butter

2.   Nintendo Switch, Nintendo Switch Case, Super Mario Odyssey

3.   Gatorade, Oatmeal, Butter, Cheerios, Milk, Tennis Racquet

4.    Dell Laptop, Wrigley’s Chewing Gum

5.   Oatmeal, Gatorade, Cheerios, Milk, Laundry Detergent

6.   Super Mario Odyssey, Red Sox Hat, Blue Jeans

7.    Oatmeal, Frozen Waffles, Butter

8.   Pepsi, Frozen Waffles, Red Sox Hat

9.   Butter, Frozen Waffles, Cheerios, Bread

10.  Nintendo Switch, Nintendo Switch Case, Frozen Waffles

11.   Dell Laptop, Nintendo Switch Case, Super Mario Odyssey

12.  Gatorade, Bread, Butter, Milk, Frozen Waffles

13.  Red Sox Hat, Blue Jeans, Oatmeal, Butter

14.  Milk, Butter, Bread

15.  Gatorade, Red Sox Hat, NIntendo Switch

1.   What is the support for {Red Sox Hat}?

2.   What is the confidence for IF {Bread} THEN {Butter}?

3.    What is the confidence for IF {Nintendo Switch} THEN {Nintendo Switch Case}?

4.    What is the lift ratio for IF {Oatmeal} THEN {Frozen Waffles}?

5.   What is the lift ratio for IF {Frozen Waffles} THEN {Milk} ?

6.  The table below shows categorical values -- a 1 in a particular cell indicates that the store carries in the item in stock, whereas a 0 indicates that the item is not stocked by that store.  Given the information contained in the table below, what is the Jacquard coefficient between Norwood and Orange?

AD699 Sporting Goods Store
Store ID	Tennis Racquets	Sunscreen	Basketball Hoops	Boogie Boards	Hawaiian Shirts
Leominster	1	0	1	0	1
Malden	1	1	1	1	1
Norwood	1	0	0	1	0
Orange	0	1	1	1	1
Pepperell	1	1	1	1	0

7.       You  recently  logged  into  Amazon.com  and  you  bought  a  video  game  called

“Splatoon.”  The next time that you logged in, you got a product recommendation based on a previous purchase.  It said, “You might be interested in Roblox.  People who bought Splatoon also bought Roblox.”

This is an example of what type of filtering?

a.   User-based collaborative filtering.

b.   Item-based collaborative filtering.

c.   Content-based filtering.

d.   Non-collaborative user filtering.

8.          ** To answer this question, a little bit of domain knowledge is required -- you just need to know that Boston is south of Cambridge **

You recently overhead a conversation on the Green Line.

“Excuse me, excuse me,” someone asked a guy wearing a backwards Red Sox cap.  “I’m from  out  of town  and  I  have  a  question  for  you  --  how  far  away  is  Boston  from Cambridge?”

The guy in the Red Sox hat said, “Sure, no problem.  They’re about 1 millimeter apart.”

“Don’t be a wiseguy,” the tourist responded.  I maybe from somewhere else, but I know that’s not true.”

The Bostonian answered with this: “Well, you never told me how you wanted to define the distance metric.  I was using the northernmost point in Boston, on the Longfellow Bridge, and giving you the distance to the southernmost point in Cambridge, which is also on that bridge.  So if you don’t like the way I answer your questions, here’s an idea -- stop asking me.”

What method did the Bostonian use to measure the distance here?

a.   Single linkage

b.   Average linkage

c.   Ward’s Method

d.   Complete linkage

9.        Take  a  look  at  the  dendrogram  shown  below,  which  depicts  an  agglomerative hierarchical clustering model.   At a distance of .25, how many clusters are there in this model?

 

10. It can be said that hierarchical agglomerative clusters are “nested.”  What does this mean?

a.   Because the initial clustering assignments in a hierarchical model are randomized, smaller clusters can often be contained inside of larger ones.

b.   If  records  have  already  formed  into  a  cluster,  they  will  remain clustered  with  those  same  records,  even  as  new  clusters  form. as higher distance cutoffs are used.

c.   As higher distance cutoffs are used, and the number of total clusters increases, there is an increasing likelihood of instability among the existing clusters.

d.   If the dataset has not been normalized, it’s possible that quirks in the data will lead to the “nesting” problem for newly-formed clusters.

11. For the graph shown below, which shows an undirected network, list all the people with a betweenness score of 0:

12.   If an  undirected network has 12 nodes, what is its maximum possible number of edges?

13. Based on the graph shown in Question 11, which of the following is true?

a.   This network is a clique, but not a connected network.

b.   This network is a connected network, but not a clique.

c.   This network is neither a clique nor a connected network.

d.   This network is both a clique and a connected network.

14.  On the web, the owner of a particular page may link to as many other web pages as he or she desires.  However, when site A links to site B, there is no requirement that site B then creates a link back to site A.  The web is which of the following?

a.   An undirected network.

b.   A conglomerated network.

c.   A directed network.

d.  An egocentric network.

15.  The lift ratio for IF {Pearl Jam} THEN {Nirvana} is 1.71.  If the support for {Nirvana} is .41, what is the confidence for IF {Pearl Jam} THEN {Nirvana}?