Scrutable & Persuasive Push-Notifications Kieran Fraser, Bilal - - PowerPoint PPT Presentation

▶

Feb 26, 2023 345 likes •674 views

Scrutable & Persuasive Push-Notifications Kieran Fraser, Bilal Yousuf, Owen Conlan ADAPT Centre, Trinity College Dublin The ADAPT Centre is funded under the SFI Research Centres Programme (Grant 13/RC/2106) and is co-funded under the

SLIDE 1

Scrutable & Persuasive Push-Notifications

Kieran Fraser, Bilal Yousuf, Owen Conlan ADAPT Centre, Trinity College Dublin

The ADAPT Centre is funded under the SFI Research Centres Programme (Grant 13/RC/2106) and is co-funded under the European Regional Development Fund.

SLIDE 2

www.adaptcentre.ie

Content

 Motivation  Problem Statement  Design  Implementation  Results  Conclusion

SLIDE 3

www.adaptcentre.ie

Motivation

http://www.stylerug.net/wp-content/uploads/2016/09/mobile-push-notification-notifyvisitors-1-1.png

Library Alert

Your textbook is now ready for collection.

Library Alert

Found a paper which could help you with your latest assignment.

Library Alert

Workshop on writing this evening.

SLIDE 4

www.adaptcentre.ie

Motivation

SLIDE 5

www.adaptcentre.ie

Motivation

SLIDE 6

www.adaptcentre.ie

Motivation

SLIDE 7

www.adaptcentre.ie

Motivation

Growing number of notifications pushed at users (Pielot, M. et al, 2014). Large no. of incoming notifications = negative user emotions (Sahami Shirazi, A. et al, 2014). Notification delivery not smart (Mehrotra, A. et al, 2016). Unnecessary notifications may dramatically decrease productivity (Iqbal, S.

T. et al, 2010).

SLIDE 8

www.adaptcentre.ie

Motivation

SLIDE 9

www.adaptcentre.ie

Motivation

SLIDE 10

www.adaptcentre.ie

Problem Statement

1. How can we make persuasive push-

notifications scrutable?

2. How

can we transparently generate persuasive push-notifications that benefit the end-user, but also maximise Click- Through-Rate?

SLIDE 11

www.adaptcentre.ie

Design

Authority (P1) People follow and respect requests made by an authority

Priority
App
Contact relevant to context

e.g. RyanAir offers at the airport

SLIDE 12

www.adaptcentre.ie

Design

Scarcity (P2) People will place higher value on something which is rare

Notification features which appear less frequently

rank higher e.g. subject, category

SLIDE 13

www.adaptcentre.ie

Design

Liking (P3) People will follow what they like

Previously liked feature content, taking the action

’opened’ as an indicator of ‘liking’

SLIDE 14

www.adaptcentre.ie

Design

Social Proof (P4) People will do what they see their peers doing

Similar notifications opened by all other users

SLIDE 15

www.adaptcentre.ie

Design

Commitment & Consistency (P5) People tend to follow through on their word and uphold behaviours associated with their own self- image

Habits toward notifications by individual user

SLIDE 16

www.adaptcentre.ie

Design

Reciprocity (P6) People feel obliged to return a favour

App last used

e.g. if content was recently consumed in an app, the user acknowledges they received value and are more likely to be persuaded to open a notification from it.

SLIDE 17

www.adaptcentre.ie

Implementation

SLIDE 18

www.adaptcentre.ie

Implementation

SLIDE 19

www.adaptcentre.ie

Implementation

SLIDE 20

www.adaptcentre.ie

Implementation

SLIDE 21

www.adaptcentre.ie

Implementation

SLIDE 22

www.adaptcentre.ie

Results

Authority (P1) People follow and respect requests made by an authority

SLIDE 23

www.adaptcentre.ie

Results

Scarcity (P2) People will place higher value on something which is rare

SLIDE 24

www.adaptcentre.ie

Results

Social Proof (P4) People will do what they see their peers doing

SLIDE 25

www.adaptcentre.ie

Results

Feature Importance Mean Decrease Impurity to identify features best when predicting CTR

SLIDE 26

www.adaptcentre.ie

Implementation

Train on Real, Test on Synthetic RMSE F1 scores differ in range 0.02 – 0.07 indicating synthetic data imitates real world data.

SLIDE 27

www.adaptcentre.ie

Results

Personalised & Persuasive Push-Notifications Using the Generative Model to create persuasive push- notifications on demand & at scale

SLIDE 28

www.adaptcentre.ie

Conclusion

Novel Contribution

1. Method
f

extracting & scrutinizing persuasiveness

push-notifications using Cialdini’s 6 principles of behaviour

2. Method
f

generating synthetic personalized & persuasive push- notifications, on-demand, and at scale

SLIDE 29

www.adaptcentre.ie

Smart Library Notifications

https://cdn5.vectorstock.com/i/1000x1000/52/99/kids-daily-routine-activities-vector-20635299.jpg

Library Alert Your textbook is now ready for collection. Library Alert Found a paper which could help you with your latest assignment. Library Alert Workshop on writing this evening.

SLIDE 30

www.adaptcentre.ie

Feasibility

SLIDE 31

www.adaptcentre.ie

Scrutable & Persuasive Push-Notifications

Content

 Motivation  Problem Statement  Design  Implementation  Results  Conclusion

Motivation

Motivation

Motivation

Motivation

Motivation

Motivation

Motivation

Problem Statement

notifications scrutable?

can we transparently generate persuasive push-notifications that benefit the end-user, but also maximise Click- Through-Rate?

Design

Authority (P1) People follow and respect requests made by an authority

e.g. RyanAir offers at the airport

Design

Scarcity (P2) People will place higher value on something which is rare

rank higher e.g. subject, category

Design

Liking (P3) People will follow what they like

’opened’ as an indicator of ‘liking’

Design

Social Proof (P4) People will do what they see their peers doing

Design

Commitment & Consistency (P5) People tend to follow through on their word and uphold behaviours associated with their own self- image

Design

Reciprocity (P6) People feel obliged to return a favour

e.g. if content was recently consumed in an app, the user acknowledges they received value and are more likely to be persuaded to open a notification from it.

Implementation

Implementation

Implementation

Implementation

Implementation

Results

Authority (P1) People follow and respect requests made by an authority

Results

Scarcity (P2) People will place higher value on something which is rare

Results

Social Proof (P4) People will do what they see their peers doing

Results

Feature Importance Mean Decrease Impurity to identify features best when predicting CTR

Implementation

Train on Real, Test on Synthetic RMSE F1 scores differ in range 0.02 – 0.07 indicating synthetic data imitates real world data.

Results

Personalised & Persuasive Push-Notifications Using the Generative Model to create persuasive push- notifications on demand & at scale

Conclusion

Novel Contribution

extracting & scrutinizing persuasiveness

push-notifications using Cialdini’s 6 principles of behaviour

generating synthetic personalized & persuasive push- notifications, on-demand, and at scale

Smart Library Notifications

Feasibility

Thank you. Questions?

Email: kieran.fraser@adaptcentre.ie