Incentive schemes for privacy-sensitive consumers

Chong Huang, Lalitha Sankar, Anand D. Sarwate

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Businesses (retailers) often offer personalized advertisements (coupons) to individuals (consumers). While proving a customized shopping experience, such coupons can provoke strong reactions from consumers who feel their privacy has been violated. Existing models for privacy try to quantify privacy risk but do not capture the subjective experience and heterogeneous expression of privacy-sensitivity. We use a Markov decision process (MDP) model for this problem. Our model captures different consumer privacy sensitivities via a time-varying state, different coupon types via an action set for the retailer, and a cost for perceived privacy violations that depends on the action and state. The simplest version of our model has two states (“Normal” and “Alerted”), two coupons (targeted and untargeted), and consumer behavior dynamics known to the retailer.We show that the optimal coupon-offering strategy for a retailer that wishes to minimize its expected discounted cost is a stationary threshold-based policy. The threshold is a function of all model parameters: the retailer offers a targeted coupon if their belief that the consumer is in the “Alerted” state is below the threshold. We extend our model and results to consumers with multiple privacy-sensitivity states as well as coupon-dependent state transition probabilities.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages358-369
Number of pages12
Volume9406
ISBN (Print)9783319255934
DOIs
StatePublished - 2015
Event6th International Conference on Decision and Game Theory for Security, GameSec 2015 - London, United Kingdom
Duration: Nov 4 2015Nov 5 2015

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9406
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other6th International Conference on Decision and Game Theory for Security, GameSec 2015
CountryUnited Kingdom
CityLondon
Period11/4/1511/5/15

Fingerprint

Incentives
Privacy
Consumer behavior
Consumer Behaviour
Model
Decision Model
Markov Decision Process
Costs
State Transition
Transition Probability
Process Model
Time-varying
Quantify
Minimise
Dependent
Industry

Keywords

  • Markov decision processes
  • Optimal policies
  • Privacy
  • Retailer-consumer interaction

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Huang, C., Sankar, L., & Sarwate, A. D. (2015). Incentive schemes for privacy-sensitive consumers. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9406, pp. 358-369). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9406). Springer Verlag. https://doi.org/10.1007/978-3-319-25594-1_21

Incentive schemes for privacy-sensitive consumers. / Huang, Chong; Sankar, Lalitha; Sarwate, Anand D.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9406 Springer Verlag, 2015. p. 358-369 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9406).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Huang, C, Sankar, L & Sarwate, AD 2015, Incentive schemes for privacy-sensitive consumers. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 9406, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9406, Springer Verlag, pp. 358-369, 6th International Conference on Decision and Game Theory for Security, GameSec 2015, London, United Kingdom, 11/4/15. https://doi.org/10.1007/978-3-319-25594-1_21
Huang C, Sankar L, Sarwate AD. Incentive schemes for privacy-sensitive consumers. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9406. Springer Verlag. 2015. p. 358-369. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-25594-1_21
Huang, Chong ; Sankar, Lalitha ; Sarwate, Anand D. / Incentive schemes for privacy-sensitive consumers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9406 Springer Verlag, 2015. pp. 358-369 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{70f5619cfbbf4d51bdba8b87d7739f49,
title = "Incentive schemes for privacy-sensitive consumers",
abstract = "Businesses (retailers) often offer personalized advertisements (coupons) to individuals (consumers). While proving a customized shopping experience, such coupons can provoke strong reactions from consumers who feel their privacy has been violated. Existing models for privacy try to quantify privacy risk but do not capture the subjective experience and heterogeneous expression of privacy-sensitivity. We use a Markov decision process (MDP) model for this problem. Our model captures different consumer privacy sensitivities via a time-varying state, different coupon types via an action set for the retailer, and a cost for perceived privacy violations that depends on the action and state. The simplest version of our model has two states (“Normal” and “Alerted”), two coupons (targeted and untargeted), and consumer behavior dynamics known to the retailer.We show that the optimal coupon-offering strategy for a retailer that wishes to minimize its expected discounted cost is a stationary threshold-based policy. The threshold is a function of all model parameters: the retailer offers a targeted coupon if their belief that the consumer is in the “Alerted” state is below the threshold. We extend our model and results to consumers with multiple privacy-sensitivity states as well as coupon-dependent state transition probabilities.",
keywords = "Markov decision processes, Optimal policies, Privacy, Retailer-consumer interaction",
author = "Chong Huang and Lalitha Sankar and Sarwate, {Anand D.}",
year = "2015",
doi = "10.1007/978-3-319-25594-1_21",
language = "English (US)",
isbn = "9783319255934",
volume = "9406",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "358--369",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Incentive schemes for privacy-sensitive consumers

AU - Huang, Chong

AU - Sankar, Lalitha

AU - Sarwate, Anand D.

PY - 2015

Y1 - 2015

N2 - Businesses (retailers) often offer personalized advertisements (coupons) to individuals (consumers). While proving a customized shopping experience, such coupons can provoke strong reactions from consumers who feel their privacy has been violated. Existing models for privacy try to quantify privacy risk but do not capture the subjective experience and heterogeneous expression of privacy-sensitivity. We use a Markov decision process (MDP) model for this problem. Our model captures different consumer privacy sensitivities via a time-varying state, different coupon types via an action set for the retailer, and a cost for perceived privacy violations that depends on the action and state. The simplest version of our model has two states (“Normal” and “Alerted”), two coupons (targeted and untargeted), and consumer behavior dynamics known to the retailer.We show that the optimal coupon-offering strategy for a retailer that wishes to minimize its expected discounted cost is a stationary threshold-based policy. The threshold is a function of all model parameters: the retailer offers a targeted coupon if their belief that the consumer is in the “Alerted” state is below the threshold. We extend our model and results to consumers with multiple privacy-sensitivity states as well as coupon-dependent state transition probabilities.

AB - Businesses (retailers) often offer personalized advertisements (coupons) to individuals (consumers). While proving a customized shopping experience, such coupons can provoke strong reactions from consumers who feel their privacy has been violated. Existing models for privacy try to quantify privacy risk but do not capture the subjective experience and heterogeneous expression of privacy-sensitivity. We use a Markov decision process (MDP) model for this problem. Our model captures different consumer privacy sensitivities via a time-varying state, different coupon types via an action set for the retailer, and a cost for perceived privacy violations that depends on the action and state. The simplest version of our model has two states (“Normal” and “Alerted”), two coupons (targeted and untargeted), and consumer behavior dynamics known to the retailer.We show that the optimal coupon-offering strategy for a retailer that wishes to minimize its expected discounted cost is a stationary threshold-based policy. The threshold is a function of all model parameters: the retailer offers a targeted coupon if their belief that the consumer is in the “Alerted” state is below the threshold. We extend our model and results to consumers with multiple privacy-sensitivity states as well as coupon-dependent state transition probabilities.

KW - Markov decision processes

KW - Optimal policies

KW - Privacy

KW - Retailer-consumer interaction

UR - http://www.scopus.com/inward/record.url?scp=84958537805&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84958537805&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-25594-1_21

DO - 10.1007/978-3-319-25594-1_21

M3 - Conference contribution

AN - SCOPUS:84958537805

SN - 9783319255934

VL - 9406

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 358

EP - 369

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -