Enhancing recommendation acceptance: Resolving the personalization–privacy paradox in recommender systems: A privacy calculus perspective

Yedi Wang, Jiaji Zhu*, Renhuai Liu, Yushi Jiang

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

The concept of recommendation acceptance intention (RAI) refers to the willingness of an individual to adopt and use recommender systems. However, contemporary recommender systems, relying solely on accuracy as an evaluation metric have demonstrated subpar performance. The evolving landscape requires the integration of multiple objectives, including relevance, diversity, novelty, and coverage. This study investigates the influence of distinct product organization formats (similar vs. related products) on consumer reactions, aiming to reduce privacy concerns while achieving objectives such as relevance and variety. Grounded in the privacy calculus theory and personalization–privacy paradox framework, this study asserts that organization formats prompt consumers to evaluate a recommender system, considering both privacy risks and personalization benefits. Findings from three experiments and one qualitative study indicate that a related product organization format effectively decreases consumer perceptions of privacy violation while enhancing their perceptions of variety compared to a similar product organization format. This result contributes to RAI and shapes consumer preferences for recommender systems. Moreover, this research distinguishes the mechanisms underlying the influence of organization formats on RAI in the context of different product types, such as durable goods and consumables. This distinction enhances implications for designing recommender systems in e-commerce.

Original languageEnglish
Article number102755
JournalInternational Journal of Information Management
Volume76
DOIs
Publication statusPublished - Jun 2024

Bibliographical note

Publisher Copyright:
© 2024 Elsevier Ltd

Fingerprint

Dive into the research topics of 'Enhancing recommendation acceptance: Resolving the personalization–privacy paradox in recommender systems: A privacy calculus perspective'. Together they form a unique fingerprint.

Cite this