How data-sharing nudges influence people's privacy preferences: A machine learning-based analysis

EAI Endorsed Transactions on Security and Safety

article

Yang Lu¹  Shujun Li²  Alex Freitas²  Athina Ioannou³ 

[1] School of Science, Technology and Health, York St John University;School of Computing, University of Kent;School of Hospitality and Tourism Management, University of Surrey

关键词: Privacy;    Nudging;    Persuasive Technology;    Data Sharing;    User Segmentation;    User Profiling;    Machine Learning;   

DOI  :  10.4108/eai.21-12-2021.172440

学科分类：口腔科学

来源: Bern Open Publishing

【 摘 要 】

INTRODUCTION: Many online services use data-sharing nudges to solicit personal data from their customersfor personalized services.OBJECTIVES: This study aims to study people’s privacy preferences in sharing different types of personaldata under different nudging conditions, how digital nudging can change their data sharing willingness, andif people’s data sharing preferences can be predicted using their responses to a questionnaire.METHODS: This paper reports a machine learning-based analysis on people’s privacy preference patternsunder four different data-sharing nudging conditions (without nudging, monetary incentives, non-monetaryincentives, and privacy assurance). The analysis is based on data collected from 685 UK residents whoparticipated in a panel survey. Their self-reported willingness levels towards sharing 23 different types ofpersonal data were analyzed by using both unsupervised (clustering) and supervised (classification) machinelearning algorithms.RESULTS: The results led to a better understanding of people’s privacy preference patterns across differentdata-sharing nudging conditions, e.g., our participants’ preferences are distributed in a space of 48 possibleprofiles more sparsely than we expected, and the unexpected observation that all the three data-sharingnudging strategies led to an overall negative effect: they led to a reduced level of self-reported willingness formore participants, comparing with the case of no nudging at all. Our experiments with supervised machinelearning models also showed that people’s privacy (data-sharing) preference profiles can be automaticallypredicted with a good accuracy, even when a small questionnaire with just seven questions is used.CONCLUSION: Our work revealed a more complicated structure of people’s privacy preference profiles,which have some dependencies on the type of data nudging and the type of personal data shared.Such complicated privacy preference profiles can be effectively analyzed using machine learning methods,including automatic prediction based on a small questionnaire. The negative results on the overall effect ofdifferent data-sharing nudges imply that service providers should consider if and how to use such mechanismsto incentivise their consumers to share personal data. We believe that more consumer-centric and transparentmethods and tools should be used to help improve trust between consumers and service providers.

【 授权许可】

CC BY