At the end of the course students will be able to:
- Understand and develop basic recommender system prototypes for (public) media
- Understand the connection between recommender systems and public values (e.g., transparency, privacy, pluriform) and how to implement these in algorithmic systems
- Understand and apply cluster analyses to media audiences
- Prototype interfaces and develop metrics that fit the goals of the underlying algorithms
- Understand and apply value sensitive design theory and methods for developing recommender systems
Assessment
- Individual assignment (presentation and technical report, 50% of the final grade): Mid-fidelity prototype of an interface that implements insights from value-sensitive design research (presentation and technical report).
- Group assignment 2 (final paper, 50% of the final grade): Building a recommender system that support public values.
Students need to pass assignment 1 (the individual assignment) with a minimum passing grade of 5.5, before they can receive a final grade in combination with assignment 2 (the group assignment) and pass the course.
A repair test requires at least a 4 for the original test.
Prerequisites
INFOMDWR Data Wrangling.
This course is for students in the master Applied Data Science only.
|
 |
|
Recommender systems are an integral part of our daily media consumption: they compile playlists on Spotify, suggest movies on Netflix, and select (news) content for personalized social media feeds on e.g., Facebook or Twitter.
In the age of information overload, recommender systems provide orientation and help users with making choices. Through data collection and statistical modelling, the underlying algorithms identify and present content that is considered most “relevant” to users.
However, recommender systems are not objective observers and/or advisors; they carry particular norms and values that their creators consciously -and unconsciously- impart during the development and deployment of algorithms.
These factors and their social impact are highly-context dependent. For example, recommender systems are often at the centre of discussions about political polarisation on digital platforms and have been associated with the reinforcement of “tunnel vision” among users by leading them into content funnels that may reduce exposure to diversity.
This course centers on the question: how can recommender systems implement public values (e.g., trust, autonomy, diversity, sustainability) ?
To approach answers and develop prototypes, students are introduced to
1) the concept of recommender systems and the connection to public values
2) value-sensitive design theory and methods (understanding the user, defining metrics, interface design), and
3) the development of basic recommender systems for (public) media (e.g., content-based, collaborative-based, and hybrid filtering).
This course approaches recommender systems from a humanities perspective; students are challenged to critically engage with data-driven technology with an explicit focus on values.
It is less “hardcore” technical but decidedly interdisciplinary with a firm grounding in the humanities/media studies.
The course has three pillars: conceptual, design, technical. Within this integrative framework, students explore the interplay between data, technology, values, and stakeholders.
Course form
Lectures, seminars.
Literature
To be announced.
|
|
|