The proposed 2-D offline evaluation strategy which accounts for the way users explore a two-dimensional recommendation carousel by concentrating on the top-left corner and then proceeding both to the right and to the bottom across different carousels.

The proposed 2-D offline evaluation strategy which accounts for the way users explore a two-dimensional recommendation carousel by concentrating on the top-left corner and then proceeding both to the right and to the bottom across different carousels.

Source publication
Preprint
Full-text available
It is common for video-on-demand and music streaming services to adopt a user interface composed of several recommendation lists, i.e. widgets or swipeable carousels, each generated according to a specific criterion or algorithm (e.g. most recent, top popular, recommended for you, editors' choice, etc.). Selecting the appropriate combination of car...

Contexts in source publication

Context 1
... this paper we propose a definition for this carousel setting highlighting the specific characteristics that distinguish both the user interface and the user behavior from the traditional single list evaluation. We also propose a 2-D strategy for the offline evaluation tailored for a carousel setting as well as a ranking metric which takes into account how users explore such interfaces (see Figure 3). We finally report experimental results indicating that accounting for the carousel interface and user behavior changes the relative performance of recommendation models such that they lead to a different conclusion on which should be the carousel to display. ...
Context 2
... this paper we propose a definition for this carousel setting highlighting the specific characteristics that distinguish both the user interface and the user behavior from the traditional single list evaluation. We also propose a 2-D strategy for the offline evaluation tailored for a carousel setting as well as a ranking metric which takes into account how users explore such interfaces (see Figure 3). We finally report experimental results indicating that accounting for the carousel interface and user behavior changes the relative performance of recommendation models such that they lead to a different conclusion on which should be the carousel to display. ...

Similar publications

Conference Paper
Full-text available
Many video-on-demand and music streaming services provide the user with a page consisting of several recommendation lists, i.e., widgets or swipeable carousels, each built with a specific criterion (e.g., most recent, TV series, etc.). Finding efficient strategies to select which carousels to display is an active research topic of great industrial...
Article
Full-text available
Many video-on-demand and music streaming services provide the user with a page consisting of several recommendation lists, i.e., widgets or swipeable carousels, each built with specific criteria (e.g., most recent, TV series, etc.). Finding efficient strategies to select which carousels to display is an active research topic of great industrial int...
Conference Paper
Full-text available
It is common for video-on-demand and music streaming services to adopt a user interface composed of several recommendation lists, i.e., widgets or swipeable carousels, each generated according to a specific criterion or algorithm (e.g., most recent, top popular, recommended for you, editors’ choice, etc.). Selecting the appropriate combination of c...
Preprint
Full-text available
Repetition in music consumption is a common phenomenon. It is notably more frequent when compared to the consumption of other media, such as books and movies. In this paper, we show that one particularly interesting repetitive behavior arises when users are consuming new items. Users' interest tends to rise with the first repetitions and attains a...