Supplemen a y Ma e ial – Su ey and In e iew Ins umen s
Su ey Ques ions
The ollowing ques ions and hei o icial esponse op ions we e p esen ed o su ey
pa icipan s.
1. Q1. Consen : Do you ag ee o ake pa in his su ey?
• Yes
• No
2. Q2. Wha is you age g oup?
• 18–24
• 25–34
• 35–44
• 45–54
• 55 o olde
3. Q3. Wha is you highes le el o educa ion?
• Bachelo
• Mas e
• PhD
• O he
4. Q4. How would you a e you academic o echnical backg ound in A i icial In elligence
(AI)?
• None (No backg ound a all)
• Basic (Hea d o AI bu ha en’ s udied i )
• In e media e (S udied o explo ed AI o lea ning o wo k)
• Ad anced (Academic o p o essional expe ience in AI)
5. Q5. How would you a e you academic o echnical backg ound in Na u al Language
P ocessing (NLP)?
• None (No backg ound a all)
• Basic (Hea d o NLP bu ha en’ s udied i )
• In e media e (S udied o explo ed NLP o lea ning o wo k)
• Ad anced (Academic o p o essional expe ience in NLP)
6. Q6. Ha e you p e iously in e ac ed wi h AI-based sys ems?
• Yes
• No
• No su e
7. Q7. Do you usually ead use e iews be o e choosing heal hca e se ices?
• Yes
• No
• Some imes
8. Q8. Wha is you p e e ed app oach when eading pa ien e iews?
• Ca e ul eading
• Skimming
• Scanning o keywo ds only
9. Q9. On a e age, how much ime do you spend eading pa ien s’ e iews be o e making a
heal hca e- ela ed decision?
• Less han 1 minu e
• 1–3 minu es
• 4–6 minu es
• 7–10 minu es
• Mo e han 10 minu es
10. Q10. How help ul do you ind eading o checking online pa ien e iews when making
heal hca e- ela ed decisions?
• 1(No help ul a all)
• 2
• 3
• 4
• 5 (Ex emely help ul)
11. Q11. To wha ex en do you ag ee wi h he ollowing s a emen s abou he use ulness o
an AI sys em ha analyses and summa ises pa ien e iews, and explains he easons
behind i s conclusions, in suppo ing heal hca e decision-making?
A. Sa es ime by expedi ing he e iew o la ge olumes o pa ien eedback.
• S ongly disag ee
• Disag ee
• Neu al
• Ag ee
• S ongly ag ee
B. Facili a es access o essen ial in o ma ion wi hou he need o ead ull e iews.
• S ongly disag ee
• Disag ee
• Neu al
• Ag ee
• S ongly ag ee
C. Suppo s decision-making based on p io i ies by o ganising commen s
acco ding o speci ic aspec s (such as wai ing ime, cleanliness, cos ).
• S ongly disag ee
• Disag ee
• Neu al
• Ag ee
• S ongly ag ee
D. Reduces bias by p o iding s anda dised analysis o e iew con en .
• S ongly disag ee
• Disag ee
• Neu al
• Ag ee
• S ongly ag ee
E. Enhances decision eliabili y h ough accu a e summa ies and clea
explana ions.
• S ongly disag ee
• Disag ee
• Neu al
• Ag ee
• S ongly ag ee
12. Q12. How impo an is i o you o unde s and why a e iew was classi ied as posi i e
o nega i e?
• 1 (No impo an a all)
• 2
• 3
• 4
• 5 (Ex emely impo an )
13. Q13. To wha ex en would knowing he eason behind he e iew’s classi ica ion
inc ease you us in he sys em?
• 1 (No a all)
• 2
• 3
• 4
• 5 (To a g ea ex en )
14. Q14. Which o he ollowing explana ion o ma s would bes help you unde s and he
sys em’s classi ica ion esul s?
• B ie ex
• Key wo ds
• G aphical elemen s
• Mixed
15. Q15. Ha e you e e used any applica ions o websi es ha analyse o explain use
e iews?
• Yes
• No
16. Q16. I you answe ed 'Yes' o he p e ious ques ion, please p o ide u he de ails
below:
(Open-ended ques ion – pa icipan s p o ided ee- ex esponses)
17. Q17. Wha aspec s should he p oposed sys em ha e o make i use ul o appealing o
you?
(Open-ended ques ion – pa icipan s p o ided ee- ex esponses)
18. Q18. Wha aspec s o he sys em migh discou age you om using i o make you
hesi an o ely on i ?
(Open-ended ques ion – pa icipan s p o ided ee- ex esponses)
19. Q19. Would you be willing o be con ac ed o a sho ollow-up in e iew o u he
discuss you iews on AI-based sys ems o heal hca e e iews?
• Yes, I’m happy o be con ac ed
• No, I’d p e e no o be con ac ed
20. Q20. I yes, please lea e you email add ess (op ional):
21. Q21. Please use he space below o sha e any addi ional hough s abou he sys em o
you expe ience wi h his su ey (op ional):
(Open-ended ques ion – pa icipan s p o ided ee- ex esponses)
In e iew Ques ions
The in e iew consis ed o he ollowing ou open-ended ques ions:
1. Q1. Wha a e he main echnical challenges you ha e pe sonally aced when de eloping
o using explainable AI (XAI) sys ems?
(Open-ended ques ion – pa icipan s p o ided ee- ex esponses)
2. Q2. Wha s a egies o app oaches do you sugges o adap ing explana ion me hods in
XAI sys ems so ha hey mee he needs and unde s anding o bo h echnical and non-
echnical use s?
(Open-ended ques ion – pa icipan s p o ided ee- ex esponses)
3. Q3. Which explana ion echniques do you ind mos in e p e able o us wo hy, and
why?
(Open-ended ques ion – pa icipan s p o ided ee- ex esponses)
4. Q4. Do you o esee any pe o mance o scalabili y challenges wi h a mul i-laye ed XAI
app oach, ha s a ing wi h an ini ial explana ion gene a ed by a s anda d me hod (e.g.,
LIME, SHAP, o simila ) and e ining i ia ex e nal pos -p ocessing?
(Open-ended ques ion – pa icipan s p o ided ee- ex esponses)
