Solutions to preference manipulation in recommender systems require knowledge of meta-preferences

TL;DR

This paper argues that addressing preference manipulation in recommender systems requires understanding and incorporating users' meta-preferences to respect autonomy and prevent manipulation.

Contribution

It introduces the concept that solutions to preference manipulation must consider meta-preferences to ensure user autonomy in recommender systems.

Findings

Preference changes can be self-induced or desired by users.

Meta-preferences influence how preferences should be managed.

Respecting meta-preferences can prevent manipulative recommendations.

Abstract

Iterative machine learning algorithms used to power recommender systems often change people's preferences by trying to learn them. Further a recommender can better predict what a user will do by making its users more predictable. Some preference changes on the part of the user are self-induced and desired whether the recommender caused them or not. This paper proposes that solutions to preference manipulation in recommender systems must take into account certain meta-preferences (preferences over another preference) in order to respect the autonomy of the user and not be manipulative.