Intervention (Hidden)
Every participant is asked to listen to a set of songs as the experiment begins. This first playlist consists of a very specific assortment of songs which are arranged in a specific order so as to elicit the participant response in relation to not only the kind of music she likes, but also the amount of variability she prefers in successive songs within a listening session. Through this first listening session, we track the participants' skipping behaviour which enables us to identify their music listening preferences. Note that for the purposes of the experiment, we limit the recommendations to the works of top 10 artists in the world based on the listening activity observed on Deezer during years of 2018-2019. Hence, for building this initial playlist, and all the remaining playlists in the experiment, we utilize the music profile of the top 10 artists such as Drake, Ed Sheeran, Coldplay, Imagine Dragons, XXXTENTACION, etc. We make this conscious choice of restricting our study to this set of 10 artists because of their popularity and the higher likelihood of participants to be already aware of their past works. Involving relatively unknown artists in our analysis may confound the eventual results due to some participants already being aware of their music and some not.
Hence, the skipping activity in the first session allows us to know whether a participant prefers to listen to a similar kind of music in her listening session or she prefers variety. At the same time, we also infer what kind of music, i.e., the genre she prefers. Hence, if she is a fan of rap, her skipping activity recorded will allow us to track that.
Before beginning the second session, the participants are randomly classified into one of the three groups which decides the kind of recommendations that will be given to them. Accordingly, we have two treatment groups and one control group:
1. Treatment Group 1 is given a new set of songs based on an algorithm that follows the proposed recommendation methodology in our paper (of the same title). This recommendation is based on the amount of variability that a participant prefers during the listening session. Hence, for a participant who is classified into this group, we already know whether she is more inclined to be 'habituated' or 'variety seeking'. We have two separate playlists that cater to each of these distinctive tastes which is then fed to the participant in the second session based on which suits her the best. These playlists are generated from the works of the top 10 artists by the algorithm that can estimate the artistic difference between two songs using the cosine dissimilarity index as shown in the paper. Note that the recommendation methodology that we propose in our main paper requires the platform to track the listeners over multiple streaming songs and across multiple streaming sessions. This requires for listener to complete one listening session on her own accord, log out of the streaming platform and then log back in at some other time of her choosing to begin another streaming session. The method that we propose then leverages the listening activity in the focal as well as the past sessions of the listener to make an informed recommendation. The scope of this experiment does not cover such long tracking of participants due to several budgetary and practical reasons. Hence, we have adopted our recommendation algorithm such that it only takes into account the dynamics of the focal session without utilizing the past session data, which in this case does not exist. As a result of this, we expect this to be a factor that may make the results that we get in this analysis weaker than we expect since an essential element of our approach is untraceable and effectively unavailable in this study.
2. Treatment Group 2 gets recommended songs using content-based collaborative filtering technique that is the status-quo in the music streaming industry. Here, the recommendation is more concerned with tracking the kind of music, for example, genre, that the participant prefers. Hence, the first session allows us to learn whether the participant likes a certain kind of music, here: rap (for example, music from Drake, XXXTENTACION, Eminem, etc.) and pop-rock (for example, music from Coldplay, Ed Sheeran, etc.). With this understanding, we then feed the participants in treatment group 2 a playlist that best reflects their musical taste between rap music and pop-rock music. These playlists are also randomly generated from the songs released by the top 10 artists by an algorithm that seeks to classify songs based on their genres (acoustic attributes).
3. The control group is given a new set of randomly selected songs. These songs are randomly selected from the options available out of the total songs released by the top10 artists considered in our study.
The group assignment between the first and the second listening sessions is the only intervention designed for the present study.