Deck 13: AI Social Media Analytics for Health Systems: Understanding Consumers Preferences on Healthcare Services

Question 1

What could an ideal AI system do?

A) Perfectly mimic the human mind
B) Be able to make free choices
C) Work independently of its original purpose
D) Write health IT code

Answer 1

A

Question 2

Which of the following is not part of the AI models used in health IT?

A) k-NN
B) NN
C) Bayesian network
D) NLP

Answer 2

D

Question 3

What can data mining of social media sites help health providers to do?

A) Avoid lawsuits
B) Understand patient satisfaction
C) Advertise services
D) Predict future demand for services

Answer 3

B

Question 4

What is a function of SMA in health IT?

A) ACA
B) NN
C) CRM
D) Forfeiture

Question 5

A physician who uses a service that applies AI to comments on social media sites can then do what?

A) Adjust services offered
B) Defend against criticism
C) Argue for increased reimbursements
D) Decide to change specialties

Question 6

Emotional content in social media reviews of hospitals can be most directly analyzed using which technique?

A) Text analysis
B) Sentiment analysis
C) Relation extraction
D) NLP

Question 7

If you are a physician who receives a dashboard report from your hospital's AI workers, this will tell you:

A) how well consumers like your work.
B) what your administration likes about your work.
C) how much business to expect in the coming 6 months.
D) how your colleagues view your work.

Question 8

What are symbolic AI methods used to find about a particular physician?

A) Satisfaction conclusions
B) Potential lawsuits
C) New consumers
D) Administrative opinions

Question 9

The Zocdoc case exemplifies what kind of analysis?

A) Physician-dominated
B) Patient-centric
C) One involving NLP preprocessing
D) One using k-NN

Question 10

From where did the information in the Zocdoc case originate?

A) Europe
B) United States and Asia
C) Australia
D) Canada

Question 11

The Singapore Medical Council helps patients search for providers by which of the following identifiers?

A) Name and geography
B) Geography and ethnic group
C) Name and cost
D) Cost and ethnic group

Question 12

Which of the following sites would be most useful for finding a physician in San Francisco?

A) Zocdoc
B) Mydoc
C) Finddoc
D) Helpdoc

Question 13

In the Zocdoc case, which word related to procedures was in the knowledgebase?

A) Suture
B) Yell
C) Charge
D) Upset

Question 14

During NLP of a social media post, punctuation marks are discarded.

Question 15

Parsing trees place extracted pieces of text into a syntactic structure.

Question 16

AI analysis of a physician review would be much harder without a knowledge base.

Question 17

A knowledge base is what consumers perceive a health provider to know.

Question 18

Tokens correspond to named entities.

Question 19

Ontology rules flow from a production database.

Question 20

The use of business intelligence tools follows preprocessing and processing such as NLP in health AI.

Question 21

Mydoc.asia is a site based in Singapore.

Question 22

If you need a doctor in Beijing, you can use Finddoc.

Question 23

Zocdoc is a U.S. site for reviewing physicians.

Question 24

AI analysis of health consumer activity online can help the consumer to find _________ services.

Question 25

NLP can be used to _________ social media posts.

Question 26

Why is social media analytics (SMA) an appropriate means of understanding consumers' preferences on e-health services? Compare and contrast SMA with the survey method.

Question 27

Discuss how artificial intelligence can improve SMA. Discuss and illustrate some of the major challenges in AI-enhanced SMA.

Question 28

How and why are visualization tools useful in identifying consumer preferences?

Question 29

Give one example on the use of SMA in digital health. What are the challenges and difficulties of using SMA in digital health?

Question 30

What are key performance indicators (KPIs)? Why may their use be important for information displayed on a dashboard?

Question 31

What are the potential breakthroughs of digital health in the age of artificial intelligence?