From Ranked Lists to Carousels: A Carousel Click Model

TL;DR

This paper introduces a carousel click model to explain user interaction with carousel recommendation interfaces, demonstrating that such interfaces enable users to examine more items and improve engagement over traditional ranked lists.

Contribution

The paper presents a novel generative carousel click model and provides analytical and empirical evidence of its effectiveness in enhancing recommendation engagement.

Findings

Users examine more items in carousel models than in ranked lists.

Carousel-based recommenders achieve higher user engagement.

The model performs well with matrix factorization-based recommenders.

Abstract

Carousel-based recommendation interfaces allow users to explore recommended items in a structured, efficient, and visually-appealing way. This made them a de-facto standard approach to recommending items to end users in many real-life recommenders. In this work, we try to explain the efficiency of carousel recommenders using a \emph{carousel click model}, a generative model of user interaction with carousel-based recommender interfaces. We study this model both analytically and empirically. Our analytical results show that the user can examine more items in the carousel click model than in a single ranked list, due to the structured way of browsing. These results are supported by a series of experiments, where we integrate the carousel click model with a recommender based on matrix factorization. We show that the combined recommender performs well on held-out test data, and leads to higher engagement with recommendations than a traditional single ranked list.