Measuring Price Discrimination and Steering on E-commerce Web Sites
Publié le 6 juillet 2018
Today, many e-commerce websites personalize their content, including Netflix (movie recommendations), Amazon (prod uct suggestions), and Yelp (business reviews). In many cases, personalization provides advantages for users: for ex ample, when a user searches for an ambiguous query such as “router,” Amazon may be able to suggest the woodworking tool instead of the networking device. However, personaliza tion on e-commerce sites may also be used to the user’s dis advantage by manipulating the products shown (price steer ing) or by customizing the prices of products (price discrim ination). Unfortunately, today, we lack the tools and tech niques necessary to be able to detect such behavior.
In this paper, we make three contributions towards ad dressing this problem. First, we develop a methodology for accurately measuring when price steering and discrimina tion occur and implement it for a variety of e-commerce web sites. While it may seem conceptually simple to detect dif ferences between users’ results, accurately attributing these differences to price discrimination and steering requires cor rectly addressing a number of sources of noise. Second, we use the accounts and cookies of over 300 real-world users to detect price steering and discrimination on 16 popular e-commerce sites. We find evidence for some form of per sonalization on nine of these e-commerce sites. Third, we investigate the effect of user behaviors on personalization. We create fake accounts to simulate different user features including web browser/OS choice, owning an account, and history of purchased or viewed products. Overall, we find numerous instances of price steering and discrimination on a variety of top e-commerce sites.
CATÉGORIESTechniques algorithmiques Types de systèmes algorithmiques Apprentissage (machine learning): supervisé (prédictif) ou non-supervisé Domaines d'application Finance Explicabilité des algorithmes Jeux de données Marketing et Publicité Modes de production: analyses d’algorithmes, black box versus white box, génération par l’algorithme, modèles d'apprentissage génératifs etc. Questions sociales, éthiques et juridiques Protection de la vie privée Provenance des données et contrôle d'usage des données Recommandation